From MAILER-DAEMON Sun Jun 01 05:01:28 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19MOiP-00041T-KJ
	for mharc-gzz-dev@gnu.org; Sun, 01 Jun 2003 05:01:21 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19MOiD-0003PR-5i
	for gzz-dev@mail.freesoftware.fsf.org; Sun, 01 Jun 2003 05:01:09 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19MOhX-0002N5-Jn
	for gzz-dev@mail.freesoftware.fsf.org; Sun, 01 Jun 2003 05:00:29 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19MOhT-0001rB-8O
	for gzz-dev@mail.freesoftware.fsf.org; Sun, 01 Jun 2003 05:00:23 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5190AaL008544;
	Sun, 1 Jun 2003 12:00:13 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h51909vb015923;
	Sun, 1 Jun 2003 12:00:09 +0300 (EETDST)
Date: Sun, 1 Jun 2003 12:00:09 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
Message-ID: <20030601090009.GJ14255@tarzan.it.jyu.fi>
References: <E19LkRU-0005Kc-0K@subversions.gnu.org> <3ED766DF.40505@gmx.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3ED766DF.40505@gmx.de>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Subject: [Gzz] Re: [Gzz-commits] manuscripts/storm short-paper.rst
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sun, 01 Jun 2003 09:01:19 -0000

On Fri, May 30, 2003 at 04:12:47PM +0200, Benja Fallenstein wrote:
> Tuomas J. Lukka wrote:
> >+One of the key innovations in storm is the use 
> >+of *pointer blocks*
> >+to build mutable documents on top of
> >+the lowest-layer append-and-delete-only model.
> >+A pointer block is simply a storm block which says
> >+``The identifier X points to document Id, overriding 
> >+the previous pointer blocks a, b`` and contains
> >+a digital signature of the above information.
> 
> I understood from Eric's comments that the terms mutable/immutable 
> aren't generally understood in connection with documents ('I didn't know 
> that immutability had something to do with versioning', or so).

Ok, once we have pointer versioning and at least text merging
properly (preferably also xu vstream text merging, as it's better),
we should write a journal article about these things.

	Tuomas



From MAILER-DAEMON Tue Jun 03 05:15:46 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19N7sn-0005Se-HL
	for mharc-gzz-dev@gnu.org; Tue, 03 Jun 2003 05:15:05 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19N7sE-00053Q-Hl
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 03 Jun 2003 05:14:30 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19N7rz-0004Bk-PE
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 03 Jun 2003 05:14:26 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19N7rx-00043j-IP
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 03 Jun 2003 05:14:13 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h539EBdc030724	for <gzz-dev@mail.freesoftware.fsf.org>;
	Tue, 3 Jun 2003 12:14:12 +0300
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1054631651.28482.15.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 03 Jun 2003 12:14:11 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti2.jyu.fi id
	h539EBdc030724
Subject: [Gzz] PEG: Storm P2P simulations using the GISP protocol
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 03 Jun 2003 09:15:03 -0000

Hi,

This is my first PEG proposal. Please feel free to comment and propose
new hypothesis or issues :).


-Hermanni


- - - - -=20

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D
PEG storm_gisp_simulation--hemppah: Storm P2P simulations using the GISP
protocol
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D

:Authors:  Hermanni Hyyti=E4l=E4
:Date-Created: 2003-06-02
:Last-Modified: $Date: 2003/06/03 09:12:19 $
:Revision: $Revision: 1.3 $
:Status:   Incomplete

.. :Stakeholders:
.. :Scope:    Major|Minor|Trivial|Cosmetic
.. :Type:     META|Policy|Architecture|Interface|Implementation

.. Affect-PEGs:


Performing GISP P2P simulations with Storm we can increase our
understanding=20
about GISP's scalability properties and on the other hand, possible
issues
related to scalability. Also, we want to know how GISP outperforms
against different threats such as network partition or security=20
attacks.

This PEG discusses how these simulations should be planned, what are
the research problems, hypothesis and issues.

Plan
=3D=3D=3D=3D

First of all, we will create a PEG document (this document) which=20
discusses general aspects of the simulation process. Then, we plan=20
to program (rather short) test cases which will test the GISP/Storm=20
P2P properties, as discussed in this document. Finally, we will=20
collect and analyse test cases' information and use this information
in the future in our manuscripts.

We plan to perform all simulations on a single computer using the
local loopback network interface to communicate with each other. The
purpose is that Storm-servers would act as if they were on different
machines. The simulation is ran under a standard Linux/Java environment.

Research problems
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

By using simulation as a research method, we try to test different kinds
of
properties of the GISP protocol without having to deploy real life
experiments.=20
There are number of research problems which we try to solve (or
understand=20
better) using the simulation process:

- How well GISP can scale if there are lot of concurrent peer joins and
  leaves in the system ? What about lookup effieciency when the network
  grows ?
        =20
- How well GISP is able to perform in adverse conditions, e.g., a
  network partition occurs ?
  =20
- How well GISP is able to perform against different kind of
  security attacks and what are the impacts ?
  =20
  =20
For now, we assume that simulation network is rather optimal, e.g.,
there
are no network latencies in the simulation network. In the future,
however,=20
we plan to perform simulations in a non-optimal network.

Hypothesis
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

- GISP can scale rather well when peers join and leave the system at a=20
  constant/static rate for a given time period and cost of
joining/leaving=20
  is logarithmic (e.g. Start with 1000 blocks and 1000 Storm-servers, 10
  peer(s) joins/leaves every 5 seconds).
  =20
- GISP can scale well and is adaptable if the cost of join/leave is=20
  logarithmic when peers join and leave the system constantly
  and the variable rate for joining/leaving changes greatly (e.g., Start
with 1000=20
  blocks and 1000 Storm-servers. 1-10 peer(s) joins/leaves every 1-10
second(s),=20
  at a given time suddenly 100-900 peers joins/leaves randomly).
  =20
- GISP's data lookup is efficient if the number of of lookup length
grows with a=20
  logarithmic growth inspite that the number of Storm-servers increases=20
  linearly (e.g. 10-10000 Storm-servers, 10000 Storm blocks, with
10-10000
  Storm-servers perform 10000 lookups randomly)
    =20
- A GISP peer is not able to handle all request properly when great
amount=20
  of query requests are performed towards a single peer/few peers (a
peer is=20
  responsible for a given key). Thus, there can be query/routing
hotspots
  in the system and load balancing properties may not scalable/tolerance
  against a hostile attack (e.g., 1000 Storm-server system, each server=20
  hosting 1-10 Storm block(s), 1-900 peers (randomly chosen) queries a=20
  single key every 1-10 second(s); calculate average block request=20
  failure, average lookup length, number of timed-out lookups and
  the distribution of lookup messages processed per peer).
    =20
- GISP is is rather fault-tolerant if 80% of lookups are succesful when
20% of
  peers die (This is Chord's simulation result) (e.g., 1000 Storm blocks
are=20
  insterted into a 1000 Storm-server system. After insertions, 1-99% of=20
  servers die randomly or in a controlled way. Before GISP starts
rebuilding=20
  routing tables, perform 1000 Storm block fetches; calculate average
block=20
  request failure, average lookup length and number of timed-out
lookups).
=20
- A hostile entity is able to reroute a data lookup to a incorrect=20
  destination peer during a data lookup process (e.g., e.g., 1000 Storm=20
  blocks are insterted into a 1000 Storm-server system in which a a
fraction
  of peers are hostile. Perform data lookups 1000 lookups randomly so
that=20
  in every lookup process, one forwarding request is rerouted
incorrectly towards
  randomly chosen destionation peer; calculate average block request
failure,=20
  average lookup length, number of timed-out lookups and the
distribution of=20
  lookup messages processed per peer).
 =20
 =20
Issues
=3D=3D=3D=3D=3D=3D

- How many virtual peers and Storm blocks we are able to simulate on a
single=20
  machine (e.g., with 256Mb of memory) ?

- In the future, do we want to perform simulations in a LAN cluster (or=20
  relevant) ?
 =20
    RESOLVED: Yes, if we want to simulate huge virtual networks and/or
    memory requirements are too massive for a single desktop.
  =20
- For now, do we need "real" Storm blocks during simulations or not ?
=20
    RESOLVED: No, since we want to make our simulator environment=20
    as lightweight as possible (See issue #1). In the future, however,
it is=20
    possible that we use "real" Storm blocks instead of "number" blocks.


Changes
=3D=3D=3D=3D=3D=3D=3D

We will program simulation test cases into the Storm CVS module.
Currently,
no changes are required to the Storm implementation codebase.





From MAILER-DAEMON Tue Jun 03 05:54:02 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19N8U0-0002QN-2p
	for mharc-gzz-dev@gnu.org; Tue, 03 Jun 2003 05:53:32 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19N8Tt-0002Jr-7c
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 03 Jun 2003 05:53:25 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19N8TO-0001WU-9N
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 03 Jun 2003 05:52:56 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19N8TH-0001FH-Hf
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 03 Jun 2003 05:52:47 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h539qeaL027126;
	Tue, 3 Jun 2003 12:52:40 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h539qca0009221;
	Tue, 3 Jun 2003 12:52:38 +0300 (EETDST)
Date: Tue, 3 Jun 2003 12:52:38 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] PEG: Storm P2P simulations using the GISP protocol
Message-ID: <20030603095238.GA6458@tarzan.it.jyu.fi>
References: <1054631651.28482.15.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1054631651.28482.15.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h539qeaL027126
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 03 Jun 2003 09:53:30 -0000

On Tue, Jun 03, 2003 at 12:14:11PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D
> PEG storm_gisp_simulation--hemppah: Storm P2P simulations using the GIS=
P
> protocol
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D
>=20
> :Authors:  Hermanni Hyyti=E4l=E4
> :Date-Created: 2003-06-02
> :Last-Modified: $Date: 2003/06/03 09:12:19 $
> :Revision: $Revision: 1.3 $
> :Status:   Incomplete
>=20
> .. :Stakeholders:
> .. :Scope:    Major|Minor|Trivial|Cosmetic
> .. :Type:     META|Policy|Architecture|Interface|Implementation
>=20
> .. Affect-PEGs:
>=20
>=20
> Performing GISP P2P simulations with Storm we can increase our
> understanding=20
> about GISP's scalability properties and on the other hand, possible
> issues
> related to scalability. Also, we want to know how GISP outperforms
> against different threats such as network partition or security=20
> attacks.

What do you mean "with Storm"?

Language:=20
understanding about -> understanding
scalability properties and... -- the part after and is redundant.
outperforms -> performs

> This PEG discusses how these simulations should be planned, what are
> the research problems, hypothesis and issues.

Wrong order, still.

Research problems, hypotheses, the theoretical knowledge we have
about the hypotheses, and possible simulations to validate=20
hypotheses.=20

Issues (PEG issues) are implicit.

> Plan
> =3D=3D=3D=3D
>=20
> First of all, we will create a PEG document (this document) which=20
> discusses general aspects of the simulation process. Then, we plan=20
> to program (rather short) test cases which will test the GISP/Storm=20
> P2P properties, as discussed in this document.=20

Nope. Still wrong order. Again: don't think about programming and simulat=
ions,
think about what questions you want answers for.

> Finally, we will=20
> collect and analyse test cases' information and use this information
> in the future in our manuscripts.

... publish any interesting discoveries.

> We plan to perform all simulations on a single computer using the
> local loopback network interface to communicate with each other.=20

Whether you need to even use the loopback depends SEVERELY on the=20
nature of the simulation -- I'd think for the first simulations=20
you will NOT need *ANY* network code.

> The
> purpose is that Storm-servers would act as if they were on different
> machines. The simulation is ran under a standard Linux/Java environment.

Umm, no. If we want to understand the theoretical properties first,
we don't want to use Storm servers as there are too many complexities the=
re
that will affect the results and make them unreadable.

> Research problems
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> By using simulation as a research method, we try to test different kind=
s
> of
> properties of the GISP protocol without having to deploy real life
> experiments.=20
> There are number of research problems which we try to solve (or
> understand=20
> better) using the simulation process:

Umm, why do you keep harping on about simulations?
Why not just say:=20

	For determining whether Storm with unmodified GISP
	is practical, we want the answers to the following questions.

If we can answer the questions without any simulations, it would be much =
better!

> - How well GISP can scale if there are lot of concurrent peer joins and
>   leaves in the system ? What about lookup effieciency when the network
>   grows ?

Isn't the "what about" part a separate question?

> - How well GISP is able to perform in adverse conditions, e.g., a
>   network partition occurs ?
>   =20
> - How well GISP is able to perform against different kind of
>   security attacks and what are the impacts ?

Ok, these sound good.=20

ISSUE: (means you need to add this to the issues section)

	Why GISP? Why are we using it versus some other systems?
	What properties does it share with others, to such a degree
	that its performance might be deduced from theirs?

ISSUE:
=09
	What experiments / simulations / methods are used in the literature?

> For now, we assume that simulation network is rather optimal, e.g.,
> there
> are no network latencies in the simulation network. In the future,
> however,=20
> we plan to perform simulations in a non-optimal network.

Don't talk about simulation even here! This is not about simulation,
this is about answering the above questions!

	At first we'll assume  ideal conditions w.r.t. ...

BTW: The correct word is ideal, not optimal: ideal is a theoretical,
unachievable simplification, but optimal is just best of the currently
existing ones.

> Hypothesis
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> - GISP can scale rather well when peers join and leave the system at a=20
>   constant/static rate for a given time period and cost of
> joining/leaving=20
>   is logarithmic (e.g. Start with 1000 blocks and 1000 Storm-servers, 1=
0
>   peer(s) joins/leaves every 5 seconds).

What do you base this hypothesis on?=20

What scales w.r.t. what?

Constant rate or constant fraction, if you scale up the number of peers?

Quite inexact still. I wouldn't know how to test this and say=20
that this is false or true.

> - GISP can scale well and is adaptable if the cost of join/leave is=20
>   logarithmic when peers join and leave the system constantly
>   and the variable rate for joining/leaving changes greatly (e.g., Star=
t
> with 1000=20
>   blocks and 1000 Storm-servers. 1-10 peer(s) joins/leaves every 1-10
> second(s),=20
>   at a given time suddenly 100-900 peers joins/leaves randomly).

Same question again.

> - GISP's data lookup is efficient if the number of of lookup length
> grows with a=20
>   logarithmic growth inspite that the number of Storm-servers increases=
=20
>   linearly (e.g. 10-10000 Storm-servers, 10000 Storm blocks, with
> 10-10000
>   Storm-servers perform 10000 lookups randomly)

"Number of of lookup length"???

> - A GISP peer is not able to handle all request properly when great
> amount=20
>   of query requests are performed towards a single peer/few peers (a
> peer is=20
>   responsible for a given key). Thus, there can be query/routing
> hotspots
>   in the system and load balancing properties may not scalable/toleranc=
e
>   against a hostile attack (e.g., 1000 Storm-server system, each server=
=20
>   hosting 1-10 Storm block(s), 1-900 peers (randomly chosen) queries a=20
>   single key every 1-10 second(s); calculate average block request=20
>   failure, average lookup length, number of timed-out lookups and
>   the distribution of lookup messages processed per peer).
>     =20
> - GISP is is rather fault-tolerant if 80% of lookups are succesful when
> 20% of
>   peers die (This is Chord's simulation result) (e.g., 1000 Storm block=
s
> are=20
>   insterted into a 1000 Storm-server system. After insertions, 1-99% of=
=20
>   servers die randomly or in a controlled way. Before GISP starts
> rebuilding=20
>   routing tables, perform 1000 Storm block fetches; calculate average
> block=20
>   request failure, average lookup length and number of timed-out
> lookups).
> =20
> - A hostile entity is able to reroute a data lookup to a incorrect=20
>   destination peer during a data lookup process (e.g., e.g., 1000 Storm=
=20
>   blocks are insterted into a 1000 Storm-server system in which a a
> fraction
>   of peers are hostile. Perform data lookups 1000 lookups randomly so
> that=20
>   in every lookup process, one forwarding request is rerouted
> incorrectly towards
>   randomly chosen destionation peer; calculate average block request
> failure,=20
>   average lookup length, number of timed-out lookups and the
> distribution of=20
>   lookup messages processed per peer).

Same problems apply to all these even more - needs to be far
more exact as to what you're claiming.

Why are you talking about 1000?

> Issues
> =3D=3D=3D=3D=3D=3D

I'd prefer to have Issues right after abstract.


> - How many virtual peers and Storm blocks we are able to simulate on a
> single=20
>   machine (e.g., with 256Mb of memory) ?

This depends on the level of abstraction - I've been trying to tell
you no real network connections are needed. If you make an abstract
enough simulation, you should be able to do millions.

> - In the future, do we want to perform simulations in a LAN cluster (or=
=20
>   relevant) ?
>  =20
>     RESOLVED: Yes, if we want to simulate huge virtual networks and/or
>     memory requirements are too massive for a single desktop.
>   =20
> - For now, do we need "real" Storm blocks during simulations or not ?
> =20
>     RESOLVED: No, since we want to make our simulator environment=20
>     as lightweight as possible (See issue #1). In the future, however,
> it is=20
>     possible that we use "real" Storm blocks instead of "number" blocks.

*WHY*?

> Changes
> =3D=3D=3D=3D=3D=3D=3D
>=20
> We will program simulation test cases into the Storm CVS module.
> Currently,
> no changes are required to the Storm implementation codebase.
>=20
>=20
>=20
>=20
> _______________________________________________
> Gzz-dev mailing list
> Gzz-dev@nongnu.org
> http://mail.nongnu.org/mailman/listinfo/gzz-dev

	TUomas



From MAILER-DAEMON Wed Jun 04 11:22:28 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Na5r-0000xG-Pi
	for mharc-gzz-dev@gnu.org; Wed, 04 Jun 2003 11:22:27 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Na5o-0000uB-4R
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 04 Jun 2003 11:22:24 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19NZyh-0006Zj-QH
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 04 Jun 2003 11:15:05 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19NZvU-0004sp-QF
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 04 Jun 2003 11:11:44 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h54FBdaL010030
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Wed, 4 Jun 2003 18:11:39 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h54FBZ65026499
	for gzz-dev@mail.freesoftware.fsf.org;
	Wed, 4 Jun 2003 18:11:35 +0300 (EETDST)
Date: Wed, 4 Jun 2003 18:11:35 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Message-ID: <20030604151135.GB21111@tarzan.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Subject: [Gzz] Hemppah's PEG round 2 comments
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 04 Jun 2003 15:22:25 -0000

>     
> ISSUE: 
> 
>     Why GISP? Why are we using it versus some other systems?
>     What properties does it share with others, to such a degree
>     that its performance might be deduced from theirs?
> 
>     RESOLVED:
>     
>     Why GISP ? Why are we using it versus some other systems?
>     
>     - According to Benja, "...since it's written in Java" :)
>     - Easy implementation (and it's already implemented)
>     - (belief that) GISP is a Kademlia_ implementation (which
>       is not true - only distance function is similar)
>      
>     It is good to mention that there is also a Python 
>     implemenation of Kademlia called Kashmir_.
>      
>     What properties does it share with others, to such a degree
>     that its performance might be deduced from theirs?
>     
>     GISP's routing table is based on Chord_'s routing 
>     table -- O(log^2 n) messages are required to join/leave 
>     operations and O(log n) for lookup efficiency (according to 
>     the original Chord publication). GISP extends Chord's routing
>     table to have more space for cached data (both the protocol
>     specification and the original publication talks about 
>     "peer information"). Chord's routing table maintains 
>     information about O(log n) peers. 
>     
>     GISP publication does not (nor the protocol specification) 
>     describe how much more space a GISP peer maintain compared 
>     to Chord's O(log n). Instead, the protocol specification 
>     states that "A peer should keep as much other
>     peer information as possible for cache". Neither 
>     the original GISP publication nor the GISP
>     protocol specification do not provide any formal
>     properties of the GISP lookup protocol. However,    
>     in the publication the author states that GISP is more efficient
>     than a "broadcasting system".    

Good, this is sounding much more like it.

ISSUE: what is known or easily deducible about the effect of a single
hostile node? Following variants:

    - Node just wants to make the network work slower

    - Node wants to make searches for some particular informations difficult

    - Node wants to make life difficult for some particular nodes,
      slowing down their queries

    - Node wants to make search for particular information by particular nodes
      difficult

    - Node wants to spam a particular identifier or xu block

What about a constant fraction of nodes?

These questions are of utmost importance to us.

ISSUE: Do we want to keep GISP or move to Kademlia or some other?

> ISSUE:
>         
>     What experiments / simulations / methods are used in the literature?
>     
>     RESOLVED:
>     
>     - mainly theoretical/best case experiments/simulations
>     - small amount of uncontrolled (real life), great amount of
>       controlled (simulation environment)
>     - simulations have been rather static and are often favored the 
>       proposed algorithm    
>     - in "real life" experiments, there have been very fast connections 
>       between peers
>     - some form of formal analyses ("proofs") are used, 
>       e.g., with Chord and Kademlia
>     - distribution of number of peers used in simulations: from 2^7 to 2^20 

A summary of what the actual findings are would be good, i.e. which
things work and how. 

Any hostile-node stuff?

> Plan
> ====
> 
> First of all, we will create a PEG document (this document) which 
> discusses general and theoretical aspects of the research problems. Then, 
> if necessary, we program (rather short) test cases which will test 
> the GISP/Storm P2P properties, as discussed in this document. Finally, we will 
> collect and analyse test cases' information and publish any interesting 
> discoveries. 
> 
> The GISP protocol
> =================
> 
> According to Daishi Kato, the author of GISP, GISP ''...intends leverage 
> those (DHT) algoritms and make a practical protocol.''. GISP's distance 
> function is based on the XOR-metric, which is similar to Kademlia's XOR-metric 
> (unsigned integer of XOR). The GISP protocol specification paper describes the 
> XOR-metric but the original GISP publication describes only numerical 
> metric (which is little confusing since Chord uses numerical distance
> metric and it's evident that a numerical metric is different from a XOR-metric).
> 
> GISP extends Chord's routing table to have more peer information as a cache.
> Thus, log^2 messages are required to join/leave operations (according to the
> original Chord publication). Additionally, Chord's routing table is asymmetric 
> (requires stabilization protocol) and lookup process is unidirectional 
> (in a virtual overlay ring). Original Kademlia publication states that 
> Chord's routing table is rigid compared to Kademlia's routing table.
> 
> Since GISP lacks of Kademlia's binary-tree-like abstraction of the routing 
> table, it is evident that the benefits of Kademlia's lookup properties (over 
> other DHTs) are not available in the GISP protocol, e.g., no concurrent and
> asynchronous lookups (no free of choice) and when a peer leaves the system
> no messages are required.

Uhh, sentence unclear: in GISP, are messages required or not?

If they are, then we should move away from it immediately!

> Original GISP publication describes "peer strength" as a measure for peer 
> heteregeneity but do not describe what properties are used and how this
> value is calculated.
> 
> For network communication the GISP protocol uses XML-RPC.
> 
> Research problems
> =================
>  
> For determining whether Storm with unmodified GISP is practical, we want the 
> answers to the following questions.
> 
> - How well GISP can scale if there are lot of concurrent peer joins and
>   leaves in the system ? 
>   
> - What about GISP's lookup effieciency when the network grows ?
> 
> - GISP does not use Kademlia's binary-tree abstraction - does it have influences 
>   and if it does what are the influences ? 

does it... --> good or bad?

> - How much better/worse pure Kademlia implementation (e.g. Kashmir) is over the GISP 
>   protocol in face performance and fault-tolerance ?

in face -> w.r.t.

> - How well GISP is able to perform in adverse conditions, e.g., a
>   network partition occurs ?
>    
> - How well GISP is able to perform against different kind of
>   security attacks and what are the impacts ?
>    
> 
> Simulations
> ===========
> 
> If needed, using a simulation process we try solve research problems. Also, 
> with a simulation process we are able test the GISP protocol without having 
> to deploy real life experiments.

Not protocol - the core algorithms.

> During simulation process, we assume that simulation network is rather ideal, 
> e.g., there are no network latencies in the simulation network. In the future,
> however, we plan to perform simulations in a non-ideal network.
> 
>    
> Hypothesis
> ==========
> 
> In this section we introduce few hypothesis related to research problems.
> Hypothesis' claims (goodness/badness) are based on the features of other DHTs 
> and their simulation results (which can be found from the literature).
> Also, values presented here (e.g., 1000-10000) are widely used in DHTs' 
> simulation processes. 
> 
> 
> - GISP's overlay can scale rather well when peers join and leave the system at a 
>   constant rate or constant fraction (both are used) for a given time period 
>   and cost of joining/leaving is logarithmic (e.g. Start with 1000 blocks and 
>   1000 Storm-servers, 10 peer(s) joins/leaves every 5 seconds).

Umm, the 1000 blocks.. example does not say what you expect the result to be
and why!

You should have a concrete estimate, including how many queries are made,
how many packets used in the network, how much delay for queries, ...

logarithmic w.r.t. what?

What's the difference between the peers sending "I'm leaving" messages
and just taking the network cable out of the wall?

Needs a LOT of work to make more exact.


> - GISP's overlay can scale well and is adaptable if the cost of join/leave is 
>   logarithmic when peers join and leave the system constantly
>   and the variable rate for joining/leaving changes greatly (e.g., Start with 1000 
>   blocks and 1000 Storm-servers. 1-10 peer(s) joins/leaves every 1-10 second(s), 
>   at a given time suddenly 100-900 peers joins/leaves randomly).
>    
> - GISP's data lookup is efficient and can scale well if lookup length grows with a 
>   logarithmic growth inspite that the number of Storm-servers increases 
>   linearly (e.g. 10-10000 Storm-servers, 10000 Storm blocks, with 10-10000
>   Storm-servers perform 10000 lookups randomly)
>      
> - A there can be query/routing hotspots in the system and load balancing properties 
>   may not scalable/tolerance against a hostile attack if a GISP peer is not able
>   to handle all request (a peer is responsible for a given key) (e.g., 1000 
>   Storm-server system, each server hosting 1-10 Storm block(s), 1-900 peers 
>   (randomly chosen) queries a single key every 1-10 second(s); calculate average 
>   block request failure, average lookup length, number of timed-out lookups and
>   the distribution of lookup messages processed per peer).
>      
> - We can say that GISP is is rather fault-tolerant if 80% of lookups are succesful 
>   when 20% of peers die (this is Chord's simulation result) (e.g., 1000 Storm blocks are 
>   insterted into a 1000 Storm-server system. After insertions, 1-99% of 
>   servers die randomly or in a controlled way. Before GISP starts rebuilding 
>   routing tables, perform 1000 Storm block fetches; calculate average block 
>   request failure, average lookup length and number of timed-out lookups).

> - A hostile entity is able to reroute a data lookup to a incorrect 
>   destination peer during a data lookup process (e.g., e.g., 1000 Storm 
>   blocks are insterted into a 1000 Storm-server system in which a a fraction
>   of peers are hostile. Perform data lookups 1000 lookups randomly so that 
>   in every lookup process, one forwarding request is rerouted incorrectly towards
>   randomly chosen destionation peer; calculate average block request failure, 
>   average lookup length, number of timed-out lookups and the distribution of 
>   lookup messages processed per peer).  

Also, hostile node can direct towards non-peers, or empty network spots
where it'll take a long time for the query to time out.

This is the most interesting part.

    Tuomas



From MAILER-DAEMON Wed Jun 04 18:08:40 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19NgLt-0004Ap-4g
	for mharc-gzz-dev@gnu.org; Wed, 04 Jun 2003 18:03:25 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19NgJA-0002yy-4F
	for gzz-dev@nongnu.org; Wed, 04 Jun 2003 18:00:36 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19NgHx-0002Fc-L0
	for gzz-dev@nongnu.org; Wed, 04 Jun 2003 17:59:22 -0400
Received: from imap.gmx.net ([213.165.64.20] helo=mail.gmx.net)
	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19NgHW-0001zU-Go
	for gzz-dev@nongnu.org; Wed, 04 Jun 2003 17:58:54 -0400
Received: (qmail 24516 invoked by uid 65534); 4 Jun 2003 21:58:51 -0000
Received: from p5082DD18.dip.t-dialin.net (EHLO gmx.de) (80.130.221.24)
  by mail.gmx.net (mp001) with SMTP; 04 Jun 2003 23:58:51 +0200
Message-ID: <3EDE6B31.2030801@gmx.de>
Date: Wed, 04 Jun 2003 23:57:05 +0200
From: Benja Fallenstein <b.fallenstein@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.3.1) Gecko/20030527 Debian/1.3.1-2
X-Accept-Language: en
MIME-Version: 1.0
To: GZZ developers <gzz-dev@nongnu.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Subject: [Gzz] Protozilla
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 04 Jun 2003 22:03:20 -0000


Hi all,

Protozilla <protozilla.mozdev.org> is a Mozilla plugin that allows you 
to create new URI schemes easily e.g. by redirection to existing URIs 
(think http). Should be usable to integrate Moz with Storm.

Unforch you would have to be root to install it if your Mozilla is in 
/usr. So on you'll have to install a copy of Mozilla in your home to use 
Protozilla securely. Annoying, but should be workable.

- Benja




From MAILER-DAEMON Thu Jun 05 00:53:50 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Nmkv-0002ol-Bi
	for mharc-gzz-dev@gnu.org; Thu, 05 Jun 2003 00:53:41 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19NmkX-0002A6-Rv
	for gzz-dev@nongnu.org; Thu, 05 Jun 2003 00:53:17 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19NmkU-00024i-K8
	for gzz-dev@nongnu.org; Thu, 05 Jun 2003 00:53:15 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19NmiU-00015O-PP
	for gzz-dev@nongnu.org; Thu, 05 Jun 2003 00:51:10 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h554p5aL024182;
	Thu, 5 Jun 2003 07:51:05 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h554oxbJ000391;
	Thu, 5 Jun 2003 07:50:59 +0300 (EETDST)
Date: Thu, 5 Jun 2003 07:50:59 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
Subject: Re: [Gzz] Protozilla
Message-ID: <20030605045059.GC21111@tarzan.it.jyu.fi>
References: <3EDE6B31.2030801@gmx.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3EDE6B31.2030801@gmx.de>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
cc: GZZ developers <gzz-dev@nongnu.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Thu, 05 Jun 2003 04:53:33 -0000

On Wed, Jun 04, 2003 at 11:57:05PM +0200, Benja Fallenstein wrote:
> 
> Hi all,
> 
> Protozilla <protozilla.mozdev.org> is a Mozilla plugin that allows you 
> to create new URI schemes easily e.g. by redirection to existing URIs 
> (think http). Should be usable to integrate Moz with Storm.

Sounds good.

> Unforch you would have to be root to install it if your Mozilla is in 
> /usr. So on you'll have to install a copy of Mozilla in your home to use 
> Protozilla securely. Annoying, but should be workable.

Really? I do think you can also put plugins in your .mozilla directory...

	Tuomas



From MAILER-DAEMON Thu Jun 05 04:23:39 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Nq1r-0005Jc-Ox
	for mharc-gzz-dev@gnu.org; Thu, 05 Jun 2003 04:23:23 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Nq0q-0004Bb-D3
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 04:22:20 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19NpzU-0003UY-RH
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 04:20:58 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Npz5-00037g-GW
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 04:20:31 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h558KJFb020712;	Thu, 5 Jun 2003 11:20:19 +0300
Subject: Re: [Gzz] Hemppah's PEG round 2 comments
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030604151135.GB21111@tarzan.it.jyu.fi>
References: <20030604151135.GB21111@tarzan.it.jyu.fi>
Content-Type: text/plain
Organization: 
Message-Id: <1054801219.1705.38.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 05 Jun 2003 11:20:19 +0300
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Thu, 05 Jun 2003 08:23:22 -0000

On Wed, 2003-06-04 at 18:11, Tuomas Lukka wrote:


> ISSUE: what is known or easily deducible about the effect of a single
> hostile node? Following variants:
> 
>     - Node just wants to make the network work slower
> 
>     - Node wants to make searches for some particular informations difficult
> 
>     - Node wants to make life difficult for some particular nodes,
>       slowing down their queries
> 
>     - Node wants to make search for particular information by particular nodes
>       difficult
> 
>     - Node wants to spam a particular identifier or xu block
> 
> What about a constant fraction of nodes?
> 
> These questions are of utmost importance to us.


Directly related to DHTs, there are two publications about this topic:

"Security Considerations for Peer-to-Peer Distributed Hash Tables"
by Emil Sit and Robert Morris

"Security for structured peer-to-peer overlay networks"
by M. Castro, P. Druschel, A. Ganesh, A. Rowstron and and D. Wallach

Both papers discusses scenarions you mentioned above, e.g., Sit's
paper describes:

- incorrect lookup routing
- incorrect routing updates
- partition
- storage and retrieval attacks
- inconsistent behavior
- overload of targeted nodes
- rapid joins and leaves
- unsolicited messages

(these are the ones that Benja listed yesterday at #fenfire)

Castro's paper discusses:

-secure nodeId assignment methods
-secure routing table maintenance
-secure message forwarding
-self-certifying data

Castro et al. first describe briefly different kinds of attacks
ans possible solutions to them.

Additionally, "Controlling the Cost of Reliability in Peer-to-Peer
Overlays" focuses DHTs' capability to self-organise in highly
adverse conditions (e.g., a network partition). The paper
is written by Ratul Mahajan, Miguel Castro and Antony Rowston.

> 
> ISSUE: Do we want to keep GISP or move to Kademlia or some other?
> 
> > ISSUE:
> >         
> >     What experiments / simulations / methods are used in the literature?
> >     
> >     RESOLVED:
> >     
> >     - mainly theoretical/best case experiments/simulations
> >     - small amount of uncontrolled (real life), great amount of
> >       controlled (simulation environment)
> >     - simulations have been rather static and are often favored the 
> >       proposed algorithm    
> >     - in "real life" experiments, there have been very fast connections 
> >       between peers
> >     - some form of formal analyses ("proofs") are used, 
> >       e.g., with Chord and Kademlia
> >     - distribution of number of peers used in simulations: from 2^7 to 2^20 
> 
> A summary of what the actual findings are would be good, i.e. which
> things work and how. 

Hm, how detailed information do you want ? Could you give an example of
"which thing" ? :)

In general, based on the published papers, it's quite hard say for sure
what things *really* work and what works "in the article".


> 
> Any hostile-node stuff?

Yes, look above.


> 
> > Plan
> > ====
> > 
> > First of all, we will create a PEG document (this document) which 
> > discusses general and theoretical aspects of the research problems. Then, 
> > if necessary, we program (rather short) test cases which will test 
> > the GISP/Storm P2P properties, as discussed in this document. Finally, we will 
> > collect and analyse test cases' information and publish any interesting 
> > discoveries. 
> > 
> > The GISP protocol
> > =================
> > 
> > According to Daishi Kato, the author of GISP, GISP ''...intends leverage 
> > those (DHT) algoritms and make a practical protocol.''. GISP's distance 
> > function is based on the XOR-metric, which is similar to Kademlia's XOR-metric 
> > (unsigned integer of XOR). The GISP protocol specification paper describes the 
> > XOR-metric but the original GISP publication describes only numerical 
> > metric (which is little confusing since Chord uses numerical distance
> > metric and it's evident that a numerical metric is different from a XOR-metric).
> > 
> > GISP extends Chord's routing table to have more peer information as a cache.
> > Thus, log^2 messages are required to join/leave operations (according to the
> > original Chord publication). Additionally, Chord's routing table is asymmetric 
> > (requires stabilization protocol) and lookup process is unidirectional 
> > (in a virtual overlay ring). Original Kademlia publication states that 
> > Chord's routing table is rigid compared to Kademlia's routing table.
> > 
> > Since GISP lacks of Kademlia's binary-tree-like abstraction of the routing 
> > table, it is evident that the benefits of Kademlia's lookup properties (over 
> > other DHTs) are not available in the GISP protocol, e.g., no concurrent and
> > asynchronous lookups (no free of choice) and when a peer leaves the system
> > no messages are required.
> 
> Uhh, sentence unclear: in GISP, are messages required or not?

Since original GISP publication talks about Chord-like routing table, we
can assume that GISP doesn't use Kademlia's routing table (abstraction).
Thus, GISP requires O(log^2 n) messages when a peer joins/leaves the
system.


> 
> If they are, then we should move away from it immediately!


Hm, do have any options ? I still recommend Kademlia ;).


> 
> > Original GISP publication describes "peer strength" as a measure for peer 
> > heteregeneity but do not describe what properties are used and how this
> > value is calculated.
> > 
> > For network communication the GISP protocol uses XML-RPC.
> > 
> > Research problems
> > =================
> >  
> > For determining whether Storm with unmodified GISP is practical, we want the 
> > answers to the following questions.
> > 
> > - How well GISP can scale if there are lot of concurrent peer joins and
> >   leaves in the system ? 
> >   
> > - What about GISP's lookup effieciency when the network grows ?
> > 
> > - GISP does not use Kademlia's binary-tree abstraction - does it have influences 
> >   and if it does what are the influences ? 
> 
> does it... --> good or bad?


No, it does not. This a bad thing.

> 
> > - How much better/worse pure Kademlia implementation (e.g. Kashmir) is over the GISP 
> >   protocol in face performance and fault-tolerance ?
> 
> in face -> w.r.t.


I don't understand this :).


> 
> > - How well GISP is able to perform in adverse conditions, e.g., a
> >   network partition occurs ?
> >    
> > - How well GISP is able to perform against different kind of
> >   security attacks and what are the impacts ?
> >    
> > 
> > Simulations
> > ===========
> > 
> > If needed, using a simulation process we try solve research problems. Also, 
> > with a simulation process we are able test the GISP protocol without having 
> > to deploy real life experiments.
> 
> Not protocol - the core algorithms.


Well, I see both the protocol and algorithm important. The algortihm (a
routing table) gives a general abstraction and properties for the 
system and the protocol defines how well the system is able to benefit
from the abstraction (e.g., this is the case with Kademlia).

> 
> > During simulation process, we assume that simulation network is rather ideal, 
> > e.g., there are no network latencies in the simulation network. In the future,
> > however, we plan to perform simulations in a non-ideal network.
> > 
> >    
> > Hypothesis
> > ==========
> > 
> > In this section we introduce few hypothesis related to research problems.
> > Hypothesis' claims (goodness/badness) are based on the features of other DHTs 
> > and their simulation results (which can be found from the literature).
> > Also, values presented here (e.g., 1000-10000) are widely used in DHTs' 
> > simulation processes. 
> > 
> > 
> > - GISP's overlay can scale rather well when peers join and leave the system at a 
> >   constant rate or constant fraction (both are used) for a given time period 
> >   and cost of joining/leaving is logarithmic (e.g. Start with 1000 blocks and 
> >   1000 Storm-servers, 10 peer(s) joins/leaves every 5 seconds).
> 
> Umm, the 1000 blocks.. example does not say what you expect the result to be
> and why!
> 
> You should have a concrete estimate, including how many queries are made,
> how many packets used in the network, how much delay for queries, ...
> 
> logarithmic w.r.t. what?
> 
> What's the difference between the peers sending "I'm leaving" messages
> and just taking the network cable out of the wall?
> 
> Needs a LOT of work to make more exact.


Yes. As I said, I didn't do much work on hypothesis yesterday ;).


> > - A hostile entity is able to reroute a data lookup to a incorrect 
> >   destination peer during a data lookup process (e.g., e.g., 1000 Storm 
> >   blocks are insterted into a 1000 Storm-server system in which a a fraction
> >   of peers are hostile. Perform data lookups 1000 lookups randomly so that 
> >   in every lookup process, one forwarding request is rerouted incorrectly towards
> >   randomly chosen destionation peer; calculate average block request failure, 
> >   average lookup length, number of timed-out lookups and the distribution of 
> >   lookup messages processed per peer).  
> 
> Also, hostile node can direct towards non-peers, or empty network spots
> where it'll take a long time for the query to time out.

In the papers, mentioned above, there are some proposals for this kind
of scenario. 

OTOH, Kademlia has a nice feature which makes Kademlia
(rather) resistant to this attack: the "free choice" feature can be used
to select peers with low latency or few network hops. Additionally, it
can be used to select on arbitrary other criteria, such as avoiding
peers that are unreachable (due to an incompletely connected underlying
network), avoiding peers that are untrusted etc.



-Hermanni




From MAILER-DAEMON Thu Jun 05 06:18:05 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19NrnS-0001Tl-I9
	for mharc-gzz-dev@gnu.org; Thu, 05 Jun 2003 06:16:38 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19NrnG-0001KH-ST
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 06:16:26 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19NrmY-0000w7-Va
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 06:15:44 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19NrmO-0000qk-8E
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 06:15:32 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h55AFRaL017170;
	Thu, 5 Jun 2003 13:15:27 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h55AFQOE003781;
	Thu, 5 Jun 2003 13:15:26 +0300 (EETDST)
Date: Thu, 5 Jun 2003 13:15:26 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] Hemppah's PEG round 2 comments
Message-ID: <20030605101526.GD21111@tarzan.it.jyu.fi>
References: <20030604151135.GB21111@tarzan.it.jyu.fi>
	<1054801219.1705.38.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1054801219.1705.38.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h55AFRaL017170
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Thu, 05 Jun 2003 10:16:32 -0000

On Thu, Jun 05, 2003 at 11:20:19AM +0300, Hermanni Hyyti=E4l=E4 wrote:
> On Wed, 2003-06-04 at 18:11, Tuomas Lukka wrote:
>=20
>=20
> > ISSUE: what is known or easily deducible about the effect of a single
> > hostile node? Following variants:
> >=20
> >     - Node just wants to make the network work slower
> >=20
> >     - Node wants to make searches for some particular informations di=
fficult
> >=20
> >     - Node wants to make life difficult for some particular nodes,
> >       slowing down their queries
> >=20
> >     - Node wants to make search for particular information by particu=
lar nodes
> >       difficult
> >=20
> >     - Node wants to spam a particular identifier or xu block
> >=20
> > What about a constant fraction of nodes?
> >=20
> > These questions are of utmost importance to us.
>=20
>=20
> Directly related to DHTs, there are two publications about this topic:
>=20
> "Security Considerations for Peer-to-Peer Distributed Hash Tables"
> by Emil Sit and Robert Morris
>=20
> "Security for structured peer-to-peer overlay networks"
> by M. Castro, P. Druschel, A. Ganesh, A. Rowstron and and D. Wallach
>=20
> Both papers discusses scenarions you mentioned above, e.g., Sit's
> paper describes:
...

Ok, since none of the defences are implemented in GISP,=20
the simulation should be postponed to see if we can make exploit programs
and get GISP proofed against them. If not, we need another platform.

So that changes the goals; we don't want to do the abstract simulations -=
-
we want to do concrete simulation of gisp clients with a fraction of
hostile exploiters and see how easy it is to stop things.=20

Therefore, I suggest postponing this PEG and starting a new one:
"Attacking GISP", which contains your plans on how you will make an explo=
it..

> > ISSUE: Do we want to keep GISP or move to Kademlia or some other?
> >=20
> > > ISSUE:
> > >        =20
> > >     What experiments / simulations / methods are used in the litera=
ture?
> > >    =20
> > >     RESOLVED:
> > >    =20
> > >     - mainly theoretical/best case experiments/simulations
> > >     - small amount of uncontrolled (real life), great amount of
> > >       controlled (simulation environment)
> > >     - simulations have been rather static and are often favored the=
=20
> > >       proposed algorithm   =20
> > >     - in "real life" experiments, there have been very fast connect=
ions=20
> > >       between peers
> > >     - some form of formal analyses ("proofs") are used,=20
> > >       e.g., with Chord and Kademlia
> > >     - distribution of number of peers used in simulations: from 2^7=
 to 2^20=20
> >=20
> > A summary of what the actual findings are would be good, i.e. which
> > things work and how.=20
>=20
> Hm, how detailed information do you want ? Could you give an example of
> "which thing" ? :)
>=20
> In general, based on the published papers, it's quite hard say for sure
> what things *really* work and what works "in the article".

Well, that's what your summary should contain: the results and
your guesses as to their generality.

> > > Plan
> > > =3D=3D=3D=3D
> > >=20
> > > First of all, we will create a PEG document (this document) which=20
> > > discusses general and theoretical aspects of the research problems.=
 Then,=20
> > > if necessary, we program (rather short) test cases which will test=20
> > > the GISP/Storm P2P properties, as discussed in this document. Final=
ly, we will=20
> > > collect and analyse test cases' information and publish any interes=
ting=20
> > > discoveries.=20
> > >=20
> > > The GISP protocol
> > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> > >=20
> > > According to Daishi Kato, the author of GISP, GISP ''...intends lev=
erage=20
> > > those (DHT) algoritms and make a practical protocol.''. GISP's dist=
ance=20
> > > function is based on the XOR-metric, which is similar to Kademlia's=
 XOR-metric=20
> > > (unsigned integer of XOR). The GISP protocol specification paper de=
scribes the=20
> > > XOR-metric but the original GISP publication describes only numeric=
al=20
> > > metric (which is little confusing since Chord uses numerical distan=
ce
> > > metric and it's evident that a numerical metric is different from a=
 XOR-metric).
> > >=20
> > > GISP extends Chord's routing table to have more peer information as=
 a cache.
> > > Thus, log^2 messages are required to join/leave operations (accordi=
ng to the
> > > original Chord publication). Additionally, Chord's routing table is=
 asymmetric=20
> > > (requires stabilization protocol) and lookup process is unidirectio=
nal=20
> > > (in a virtual overlay ring). Original Kademlia publication states t=
hat=20
> > > Chord's routing table is rigid compared to Kademlia's routing table.
> > >=20
> > > Since GISP lacks of Kademlia's binary-tree-like abstraction of the =
routing=20
> > > table, it is evident that the benefits of Kademlia's lookup propert=
ies (over=20
> > > other DHTs) are not available in the GISP protocol, e.g., no concur=
rent and
> > > asynchronous lookups (no free of choice) and when a peer leaves the=
 system
> > > no messages are required.
> >=20
> > Uhh, sentence unclear: in GISP, are messages required or not?
>=20
> Since original GISP publication talks about Chord-like routing table, w=
e
> can assume that GISP doesn't use Kademlia's routing table (abstraction).
> Thus, GISP requires O(log^2 n) messages when a peer joins/leaves the
> system.

Which is not acceptable for us - network connections will get severed &c.

So GISP doesn't even know how to deal with machines vanishing from the ne=
twork?

> > If they are, then we should move away from it immediately!
>=20
> Hm, do have any options ? I still recommend Kademlia ;).

Yes, you mentioned the python client -- might be worth looking at.



> > > - How much better/worse pure Kademlia implementation (e.g. Kashmir)=
 is over the GISP=20
> > >   protocol in face performance and fault-tolerance ?
> >=20
> > in face -> w.r.t.
>=20
>=20
> I don't understand this :).

"in face performance" is not proper English.=20
w.r.t (with respect to) is the correct term.

> >=20
> > > - How well GISP is able to perform in adverse conditions, e.g., a
> > >   network partition occurs ?
> > >   =20
> > > - How well GISP is able to perform against different kind of
> > >   security attacks and what are the impacts ?
> > >   =20
> > >=20
> > > Simulations
> > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> > >=20
> > > If needed, using a simulation process we try solve research problem=
s. Also,=20
> > > with a simulation process we are able test the GISP protocol withou=
t having=20
> > > to deploy real life experiments.
> >=20
> > Not protocol - the core algorithms.
>=20
>=20
> Well, I see both the protocol and algorithm important. The algortihm (a
> routing table) gives a general abstraction and properties for the=20
> system and the protocol defines how well the system is able to benefit
> from the abstraction (e.g., this is the case with Kademlia).

The actual implementation is not important - it can be changed.
The algorithms and the susceptibility to attacks is the big thing.

> > > - A hostile entity is able to reroute a data lookup to a incorrect=20
> > >   destination peer during a data lookup process (e.g., e.g., 1000 S=
torm=20
> > >   blocks are insterted into a 1000 Storm-server system in which a a=
 fraction
> > >   of peers are hostile. Perform data lookups 1000 lookups randomly =
so that=20
> > >   in every lookup process, one forwarding request is rerouted incor=
rectly towards
> > >   randomly chosen destionation peer; calculate average block reques=
t failure,=20
> > >   average lookup length, number of timed-out lookups and the distri=
bution of=20
> > >   lookup messages processed per peer). =20
> >=20
> > Also, hostile node can direct towards non-peers, or empty network spo=
ts
> > where it'll take a long time for the query to time out.
>=20
> In the papers, mentioned above, there are some proposals for this kind
> of scenario.=20
>=20
> OTOH, Kademlia has a nice feature which makes Kademlia
> (rather) resistant to this attack: the "free choice" feature can be use=
d
> to select peers with low latency or few network hops. Additionally, it
> can be used to select on arbitrary other criteria, such as avoiding
> peers that are unreachable (due to an incompletely connected underlying
> network), avoiding peers that are untrusted etc.

So we should start looking into the python client?

Remember also that if we use indices for transclusions, we have a fresh
problem as it becomes possible to spam...

	Tuomas



From MAILER-DAEMON Thu Jun 05 07:33:32 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Nswp-0008LX-BF
	for mharc-gzz-dev@gnu.org; Thu, 05 Jun 2003 07:30:23 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19NswZ-0008EX-IF
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 07:30:07 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Nsvz-0007uN-Kk
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 07:29:33 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19NsqK-0006Hs-Sp
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 07:23:41 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h55BNcaL024400;
	Thu, 5 Jun 2003 14:23:38 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h55BNaHQ004628;
	Thu, 5 Jun 2003 14:23:36 +0300 (EETDST)
Date: Thu, 5 Jun 2003 14:23:36 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] Hemppah's PEG round 2 comments
Message-ID: <20030605112336.GE21111@tarzan.it.jyu.fi>
References: <20030604151135.GB21111@tarzan.it.jyu.fi>
	<1054801219.1705.38.camel@io.it.jyu.fi>
	<20030605101526.GD21111@tarzan.it.jyu.fi>
	<1054810176.1705.52.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1054810176.1705.52.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h55BNcaL024400
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Thu, 05 Jun 2003 11:30:20 -0000

On Thu, Jun 05, 2003 at 01:49:36PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> > Remember also that if we use indices for transclusions, we have a fre=
sh
> > problem as it becomes possible to spam...
>=20
> Btw, there is a paper about DHT design, which is designed to be
> resilient against spam attacks:
> http://iptps03.cs.berkeley.edu/final-papers/simple_fault_tolerant.pdf

You should read these things more carefully: the spam they're talking
about is nothing like the spam I'm talking about.

	Tuomas



From MAILER-DAEMON Thu Jun 05 08:01:40 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19NtOf-0008BE-S0
	for mharc-gzz-dev@gnu.org; Thu, 05 Jun 2003 07:59:09 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19NtNl-0007ov-Jh
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 07:58:13 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19NtNE-0007c0-0K
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 07:57:40 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19NsJO-0002vK-TF
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 06:49:39 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h55AnaFb022764;	Thu, 5 Jun 2003 13:49:36 +0300
Subject: Re: [Gzz] Hemppah's PEG round 2 comments
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030605101526.GD21111@tarzan.it.jyu.fi>
References: <20030604151135.GB21111@tarzan.it.jyu.fi>
	 <1054801219.1705.38.camel@io.it.jyu.fi>
	 <20030605101526.GD21111@tarzan.it.jyu.fi>
Content-Type: text/plain
Organization: 
Message-Id: <1054810176.1705.52.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 05 Jun 2003 13:49:36 +0300
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Thu, 05 Jun 2003 11:59:08 -0000

On Thu, 2003-06-05 at 13:15, Tuomas Lukka wrote:

> > 
> > Directly related to DHTs, there are two publications about this topic:
> > 
> > "Security Considerations for Peer-to-Peer Distributed Hash Tables"
> > by Emil Sit and Robert Morris
> > 
> > "Security for structured peer-to-peer overlay networks"
> > by M. Castro, P. Druschel, A. Ganesh, A. Rowstron and and D. Wallach
> > 
> > Both papers discusses scenarions you mentioned above, e.g., Sit's
> > paper describes:
> ...
> 
> Ok, since none of the defences are implemented in GISP, 
> the simulation should be postponed to see if we can make exploit programs
> and get GISP proofed against them. If not, we need another platform.
> 
> So that changes the goals; we don't want to do the abstract simulations --
> we want to do concrete simulation of gisp clients with a fraction of
> hostile exploiters and see how easy it is to stop things. 
> 
> Therefore, I suggest postponing this PEG and starting a new one:
> "Attacking GISP", which contains your plans on how you will make an exploit..


Ok, will do that.


> > 
> > Since original GISP publication talks about Chord-like routing table, we
> > can assume that GISP doesn't use Kademlia's routing table (abstraction).
> > Thus, GISP requires O(log^2 n) messages when a peer joins/leaves the
> > system.
> 
> Which is not acceptable for us - network connections will get severed &c.
> 
> So GISP doesn't even know how to deal with machines vanishing from the network?

No, it *does* know how to deal with leaving peers -- by using same
mechanisms se Chord. But as said, Chord's routing table/virtual overlay
topology is rigid for a highly dynamic environment.

> 
> > > If they are, then we should move away from it immediately!
> > 
> > Hm, do have any options ? I still recommend Kademlia ;).
> 
> Yes, you mentioned the python client -- might be worth looking at.

Indeed. First, however, I will ask from the authors of Kashmir how
similar their implementation is compared to Kademlia paper. Hm, perhaps
I should also ask from the authors of Kademlia paper what's the current
situation of their C++ implementation of Kademlia (brief history: first,
there was a Java implementation and now they are working with the C++
implementation).



> Remember also that if we use indices for transclusions, we have a fresh
> problem as it becomes possible to spam...

Btw, there is a paper about DHT design, which is designed to be
resilient against spam attacks:
http://iptps03.cs.berkeley.edu/final-papers/simple_fault_tolerant.pdf



-Hermanni





From MAILER-DAEMON Thu Jun 05 08:42:57 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Nu4S-0006UJ-Kt
	for mharc-gzz-dev@gnu.org; Thu, 05 Jun 2003 08:42:20 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Nu4Q-0006SA-Qx
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 08:42:18 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Nu4P-0006Rr-G9
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 08:42:18 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Nu20-0005nt-Eg
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 08:39:48 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h55CdjaL029952;
	Thu, 5 Jun 2003 15:39:45 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h55CdinQ007157;
	Thu, 5 Jun 2003 15:39:44 +0300 (EETDST)
Date: Thu, 5 Jun 2003 15:39:43 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] Hemppah's PEG round 2 comments
Message-ID: <20030605123943.GC6356@tarzan.it.jyu.fi>
References: <20030604151135.GB21111@tarzan.it.jyu.fi>
	<1054801219.1705.38.camel@io.it.jyu.fi>
	<20030605101526.GD21111@tarzan.it.jyu.fi>
	<1054810176.1705.52.camel@io.it.jyu.fi>
	<20030605112336.GE21111@tarzan.it.jyu.fi>
	<1054816708.1709.95.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1054816708.1709.95.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h55CdjaL029952
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Thu, 05 Jun 2003 12:42:19 -0000

On Thu, Jun 05, 2003 at 03:38:29PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> On Thu, 2003-06-05 at 14:23, Tuomas Lukka wrote:
> > On Thu, Jun 05, 2003 at 01:49:36PM +0300, Hermanni Hyyti=E4l=E4 wrote=
:
> > > > Remember also that if we use indices for transclusions, we have a=
 fresh
> > > > problem as it becomes possible to spam...
> > >=20
> > > Btw, there is a paper about DHT design, which is designed to be
> > > resilient against spam attacks:
> > > http://iptps03.cs.berkeley.edu/final-papers/simple_fault_tolerant.p=
df
> >=20
> > You should read these things more carefully: the spam they're talking
> > about is nothing like the spam I'm talking about.
>=20
> They are talking about spam generated by a faulty peer(s). They also
> mention that their design could be resilient against spam generated by =
a
> hostile peer(s) with small design modifications.
>=20
> Which spam model are you talking about ? :)

I'm talking about one where you *insert* a lot of values into the DHT.

	Tuomas



From MAILER-DAEMON Thu Jun 05 08:47:10 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Nu97-0008Iv-LW
	for mharc-gzz-dev@gnu.org; Thu, 05 Jun 2003 08:47:09 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Nu7g-0007rF-Cc
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 08:45:40 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Nu6i-0007bh-Jp
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 08:44:41 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Nu0m-0005GY-SL
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 08:38:33 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h55CcTdc028579;	Thu, 5 Jun 2003 15:38:29 +0300
Subject: Re: [Gzz] Hemppah's PEG round 2 comments
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030605112336.GE21111@tarzan.it.jyu.fi>
References: <20030604151135.GB21111@tarzan.it.jyu.fi>
	 <1054801219.1705.38.camel@io.it.jyu.fi>
	 <20030605101526.GD21111@tarzan.it.jyu.fi>
	 <1054810176.1705.52.camel@io.it.jyu.fi>
	 <20030605112336.GE21111@tarzan.it.jyu.fi>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1054816708.1709.95.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 05 Jun 2003 15:38:29 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti2.jyu.fi id
	h55CcTdc028579
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Thu, 05 Jun 2003 12:47:04 -0000

On Thu, 2003-06-05 at 14:23, Tuomas Lukka wrote:
> On Thu, Jun 05, 2003 at 01:49:36PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> > > Remember also that if we use indices for transclusions, we have a f=
resh
> > > problem as it becomes possible to spam...
> >=20
> > Btw, there is a paper about DHT design, which is designed to be
> > resilient against spam attacks:
> > http://iptps03.cs.berkeley.edu/final-papers/simple_fault_tolerant.pdf
>=20
> You should read these things more carefully: the spam they're talking
> about is nothing like the spam I'm talking about.

They are talking about spam generated by a faulty peer(s). They also
mention that their design could be resilient against spam generated by a
hostile peer(s) with small design modifications.

Which spam model are you talking about ? :)


-Hermanni




From MAILER-DAEMON Thu Jun 05 11:37:06 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19NweZ-0001n8-Ls
	for mharc-gzz-dev@gnu.org; Thu, 05 Jun 2003 11:27:47 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19NwWI-00088i-Gv
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 11:19:14 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19NwT2-0006pV-20
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 11:15:52 -0400
Received: from tammi2.cc.jyu.fi ([130.234.4.77])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19NwQm-0006E8-SJ
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 05 Jun 2003 11:13:33 -0400
Received: (from apache@localhost)
	by tammi2.cc.jyu.fi (8.11.6/8.11.6) id h55Er7h31175
	for gzz-dev@mail.freesoftware.fsf.org; Thu, 5 Jun 2003 17:53:07 +0300
X-Authentication-Warning: tammi2.cc.jyu.fi: apache set sender to
	hemppah@cc.jyu.fi using -f
Received: from 62.142.77.125 ( [62.142.77.125])
	as user hemppah@imap.cc.jyu.fi by tammi2.cc.jyu.fi with HTTP;
	Thu,  5 Jun 2003 17:53:07 +0300
Message-ID: <1054824787.3edf5953d2af4@tammi2.cc.jyu.fi>
Date: Thu,  5 Jun 2003 17:53:07 +0300
From: hemppah@cc.jyu.fi
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
User-Agent: Internet Messaging Program (IMP) 3.1
X-Originating-IP: 62.142.77.125
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by tammi2.cc.jyu.fi id
	h55Er7h31175
Subject: [Gzz] Fwd: Re: A few questions regarding Kademlia
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Thu, 05 Jun 2003 15:27:45 -0000


----- Forwarded message from Petar Maymounkov <petar@scs.cs.nyu.edu> ----=
-
 =20
> >
> > We are currently developing a distributed hypermedia system, named
> > Fenfire (http://fenfire.org). At the lowest level, our system store d=
ata
> > as *blocks*. A block is like a file but it cannot be changed,
> > and instead of a file name it is identified by a
> > cryptographic content hash (SHA-1).
> >
> > We are building a P2P extension to our storage layer that allows one =
to
> > find and download a block if *anybody* on the 'net publishes it. We h=
ave
> > decided to use Kademlia DHT for locating data in a distributed
> > environment. However, I have a few questions regarding Kademlia:
> >
> > 1) What is the current situation with the C++ implementation of
> > Kademlia?

It is 80% ready and on hold, until the Online Codes file sharing are done.

> >
> > 2) When you plan to release the C++ implementation and would the lice=
nse
> > be GPL or LGPL, or...?

It will be available under some license that doesn't allow commercial use.

> >
> > 3) Excluding your implementation, do you know if there is a project
> > which implements fully the features which you described in the origin=
al
> > Kademlia paper?

No.

> >
> > 4) Altough GISP (http://gisp.jxta.org) implements Chord-based routing
> > table and lacks of Kademlia's routing table abstraction
> > (binary-like-tree), you list GISP one of those implements Kademlia. I=
s
> > the reason that GISP implements XOR-distance function (as Kademlia
> > does)?

We have a link to GISP only because we were told that they implement
Kademlia. We haven't checked it explicitly.

> >
> > 5) It has been over a year since the original Kademlia paper was
> > published. A lot of new DHT papers have been published thereafter.
> > Today, do you still think that Kademlia is the best DHT available ?

Yes.

> >
> >
> > Btw, Kademlia web site (http://kademlia.scs.cs.nyu.edu/) hasn't been
> > available for couple of days; Mozilla claims that the connection is
> > refused ;).

It will be down for a little while. Changing operating systems and such.
Maybe there will be new content when it comes up.

> >
> >
> > Best regards,
> > Hermanni Hyyti=E4l=E4
> >
>

Petar



----- End forwarded message -----




-------------------------------------------------
This mail sent through IMP: http://horde.org/imp/



From MAILER-DAEMON Fri Jun 06 05:50:37 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19ODpu-00086p-Pg
	for mharc-gzz-dev@gnu.org; Fri, 06 Jun 2003 05:48:38 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19ODgl-0002YH-LH
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 05:39:11 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19ODeG-0001ps-2y
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 05:36:37 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19ODdc-0001gG-SL
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 05:35:57 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h569Zrdc019868	for <gzz-dev@mail.freesoftware.fsf.org>;
	Fri, 6 Jun 2003 12:35:55 +0300
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1054892152.4089.58.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 06 Jun 2003 12:35:53 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti2.jyu.fi id
	h569Zrdc019868
Subject: [Gzz] PEG: Attacking against GISP
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 06 Jun 2003 09:48:37 -0000


=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
PEG attacking_gisp--hemppah:
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

:Authors:  Hermanni Hyyti=E4l=E4
:Date-Created: 2003-06-05
:Last-Modified: $Date: 2003/06/05 13:42:50 $
:Revision: $Revision: 1.4 $
:Status:   Incomplete

.. :Stakeholders:
.. :Scope:    Major|Minor|Trivial|Cosmetic
.. :Type:     META|Policy|Architecture|Interface|Implementation

.. Affect-PEGs:

This PEG document briefly describes the attack methods used by a
"killer"=20
program against the GISP P2P software. The program is used to test
GISP's
robustness against hostile attacks. =20

In this document we mean with "peer" as an entity which is able to do a=20
(limited/simplified/modified) number of regular GISP peer's
functionalies
in a way which may be harmful for the GISP network w.r.t. performance
and redundancy. The harmfulness of a peer be a consequence of the fact
that a peer is faulty, or a peer is wilfully malicious.=20

Disclaimer
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
This program is only used for research purposes and
the goal is to improve GISP's resilience against hostile attacks.


Research problems
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

For determining whether GISP is resilient against hostile attack,=20
we want the answers to the following questions:

- Does GISP have *obvious* exploits ?

- If there are exploits how easily they can be used by an=20
  hostile peer ?=20

- Is GISP resilient against (hostile) attacks or not ?

- How severe implications attacks may cause ?

- How well GISP is able to re-organise after a (hostile) attack ?

Simulation method
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

We will perform simulations on a single desktop computer. We intend
to use GISP's native code as much as possible. We start our=20
simulation process by creating simple scenarios in which:

- A hostile (or faulty) peer(s) doesn't reply to queries (a "dumb"
peer)*

- A hostile (or faulty) peer(s) tries to drop certain packets/queries
wilfully*
 =20
- A hostile (or faulty) peer(s) forwards queries to incorrect
destination peers

- A hostile (or faulty) peer(s) gives false information during queries

- A hostile (or faulty) peer(s)'s queries/replies include loads of
rubbish, i.e.,=20
  wrong XML-scheme, wrong string/text encoding, or doesn't otherwise
follow the protocol
 =20
- A hostile (or faulty) peer(s) performs many queries randomly

- A hostile (or faulty) peer(s) performs many queries wilfully with a
certain key

- A hostile (or faulty) peer(s) stores lot of dummy random
key-value-pairs

- A hostile (or faulty) peer(s) stores lot of dummy key-value-pairs with
a certain key

During the simulation process we will use a single hostile (or faulty)
peer
or a group of hostile/faulty peers (fraction of all peers) in the test
network.=20

By using above scenarios, we want to clarify GISP's properties with
regard to research=20
questions.

*=3DWe start the simulation process with these scenarios. =20
 =20
Issues
=3D=3D=3D=3D=3D=3D

.. Is there any issues ?

None, yet.

Changes
=3D=3D=3D=3D=3D=3D=3D

No changes are required to the Storm implementation
codebase.





From MAILER-DAEMON Fri Jun 06 07:36:38 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19OFUf-0000qw-Gv
	for mharc-gzz-dev@gnu.org; Fri, 06 Jun 2003 07:34:49 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19OFHl-0004gy-Dq
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 07:21:29 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19OF7t-0002k5-RP
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 07:11:19 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19OF3P-0001rL-NN
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 07:06:39 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h56B6UaL009290;
	Fri, 6 Jun 2003 14:06:30 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h56B6S48019950;
	Fri, 6 Jun 2003 14:06:28 +0300 (EETDST)
Date: Fri, 6 Jun 2003 14:06:28 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] PEG: Attacking against GISP
Message-ID: <20030606110628.GA18385@tarzan.it.jyu.fi>
References: <1054892152.4089.58.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1054892152.4089.58.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h56B6UaL009290
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 06 Jun 2003 11:34:47 -0000

On Fri, Jun 06, 2003 at 12:35:53PM +0300, Hermanni Hyyti=E4l=E4 wrote:
>=20
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> PEG attacking_gisp--hemppah:
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> :Authors:  Hermanni Hyyti=E4l=E4
> :Date-Created: 2003-06-05
> :Last-Modified: $Date: 2003/06/05 13:42:50 $
> :Revision: $Revision: 1.4 $
> :Status:   Incomplete
>=20
> .. :Stakeholders:
> .. :Scope:    Major|Minor|Trivial|Cosmetic
> .. :Type:     META|Policy|Architecture|Interface|Implementation
>=20
> .. Affect-PEGs:
>=20
> This PEG document briefly describes the attack methods used by a
> "killer"=20
> program against the GISP P2P software. The program is used to test
> GISP's
> robustness against hostile attacks. =20

Used -> intended to be used / tried / ...

This is a plan, not the post-factum documentation.


> In this document we mean with "peer" as an entity which is able to do a=
=20
> (limited/simplified/modified) number of regular GISP peer's
> functionalies
> in a way which may be harmful for the GISP network w.r.t. performance
> and redundancy.=20

Umm, bad choice of words - recommend "hostile peer", "faulty peer"

> The harmfulness of a peer be a consequence of the fact
> that a peer is faulty, or a peer is wilfully malicious.=20

Let's start with plain maliciousness... Anything that faultiness can do,
malice can do better.

> Disclaimer
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> This program is only used for research purposes and
> the goal is to improve GISP's resilience against hostile attacks.

.. author has stated that blah blah -- give background.

> Research problems
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> For determining whether GISP is resilient against hostile attack,=20
> we want the answers to the following questions:

Incorrect:=20

GISP does not appear to address the attacks against DHT systems
described in [...] and [...].

The question is *how* much harm can a peer do.

> Simulation method
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> We will perform simulations on a single desktop computer. We intend
> to use GISP's native code as much as possible. We start our=20
> simulation process by creating simple scenarios in which:
>=20
> - A hostile (or faulty) peer(s) doesn't reply to queries (a "dumb"
> peer)*

(but takes part in forming of the network?) ...=20

> - A hostile (or faulty) peer(s) tries to drop certain packets/queries
> wilfully*
>
> - A hostile (or faulty) peer(s) forwards queries to incorrect
> destination peers
>=20
> - A hostile (or faulty) peer(s) gives false information during queries
>=20
> - A hostile (or faulty) peer(s)'s queries/replies include loads of
> rubbish, i.e.,=20
>   wrong XML-scheme, wrong string/text encoding, or doesn't otherwise
> follow the protocol
>  =20
> - A hostile (or faulty) peer(s) performs many queries randomly
>=20
> - A hostile (or faulty) peer(s) performs many queries wilfully with a
> certain key
>=20
> - A hostile (or faulty) peer(s) stores lot of dummy random
> key-value-pairs

> - A hostile (or faulty) peer(s) stores lot of dummy key-value-pairs wit=
h
> a certain key

Does GISP allow several values per key?

> During the simulation process we will use a single hostile (or faulty)
> peer
> or a group of hostile/faulty peers (fraction of all peers) in the test
> network.=20
>=20
> By using above scenarios, we want to clarify GISP's properties with
> regard to research=20
> questions.
>=20
> *=3DWe start the simulation process with these scenarios. =20

Next you should describe the exact scenarios you want to try,
so start with the dumb peer and specify it exactly.

Then run the test.


> Issues
> =3D=3D=3D=3D=3D=3D
>=20
> .. Is there any issues ?
>=20
> None, yet.

ISSUE: When should we discuss this with the author of GISP?

	Tuomas




From MAILER-DAEMON Fri Jun 06 08:13:08 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19OG49-0004mN-7y
	for mharc-gzz-dev@gnu.org; Fri, 06 Jun 2003 08:11:29 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19OFwU-0002J3-B1
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 08:03:34 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19OFsV-0000xR-GJ
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 07:59:30 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19OFmo-0007nB-5q
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 07:53:34 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h56BrVFb007305;	Fri, 6 Jun 2003 14:53:31 +0300
Subject: Re: [Gzz] PEG: Attacking against GISP
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030606110628.GA18385@tarzan.it.jyu.fi>
References: <1054892152.4089.58.camel@io.it.jyu.fi>
	 <20030606110628.GA18385@tarzan.it.jyu.fi>
Content-Type: text/plain
Organization: 
Message-Id: <1054900410.4089.69.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 06 Jun 2003 14:53:31 +0300
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 06 Jun 2003 12:11:27 -0000

On Fri, 2003-06-06 at 14:06, Tuomas Lukka wrote:

> 
> > - A hostile (or faulty) peer(s) stores lot of dummy key-value-pairs with
> > a certain key
> 
> Does GISP allow several values per key?


Like DSHT ? No.

> 
> > During the simulation process we will use a single hostile (or faulty)
> > peer
> > or a group of hostile/faulty peers (fraction of all peers) in the test
> > network. 
> > 
> > By using above scenarios, we want to clarify GISP's properties with
> > regard to research 
> > questions.
> > 
> > *=We start the simulation process with these scenarios.  
> 
> Next you should describe the exact scenarios you want to try,
> so start with the dumb peer and specify it exactly.
> 
> Then run the test.

Ok.



-Hermanni





From MAILER-DAEMON Fri Jun 06 08:19:06 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19OGBI-000053-CX
	for mharc-gzz-dev@gnu.org; Fri, 06 Jun 2003 08:18:52 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19OGAR-0008AH-6Z
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 08:17:59 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19OGAH-00082I-TP
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 08:17:50 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19OG9d-0007We-Vh
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 06 Jun 2003 08:17:10 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h56CH6aL014430;
	Fri, 6 Jun 2003 15:17:07 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h56CH59g021442;
	Fri, 6 Jun 2003 15:17:05 +0300 (EETDST)
Date: Fri, 6 Jun 2003 15:17:05 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] PEG: Attacking against GISP
Message-ID: <20030606121705.GB18385@tarzan.it.jyu.fi>
References: <1054892152.4089.58.camel@io.it.jyu.fi>
	<20030606110628.GA18385@tarzan.it.jyu.fi>
	<1054900410.4089.69.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1054900410.4089.69.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h56CH6aL014430
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 06 Jun 2003 12:18:50 -0000

On Fri, Jun 06, 2003 at 02:53:31PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> On Fri, 2003-06-06 at 14:06, Tuomas Lukka wrote:
>=20
> >=20
> > > - A hostile (or faulty) peer(s) stores lot of dummy key-value-pairs=
 with
> > > a certain key
> >=20
> > Does GISP allow several values per key?
>=20
> Like DSHT ? No.

So how do we make it work for block and transclusion indices?

	Tuomas



From MAILER-DAEMON Sat Jun 07 14:29:05 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19OiQh-0000Wu-9e
	for mharc-gzz-dev@gnu.org; Sat, 07 Jun 2003 14:28:39 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19OiQD-0008AH-4c
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 07 Jun 2003 14:28:09 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19OiPz-0007h2-T4
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 07 Jun 2003 14:27:56 -0400
Received: from mail.gmx.de ([213.165.64.20] helo=mail.gmx.net)
	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19OiPg-0006uw-IB
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 07 Jun 2003 14:27:36 -0400
Received: (qmail 8874 invoked by uid 65534); 7 Jun 2003 18:27:30 -0000
Received: from pD9E1270B.dip.t-dialin.net (EHLO gmx.de) (217.225.39.11)
  by mail.gmx.net (mp006) with SMTP; 07 Jun 2003 20:27:30 +0200
Message-ID: <3EE22E64.9020802@gmx.de>
Date: Sat, 07 Jun 2003 20:26:44 +0200
From: Benja Fallenstein <b.fallenstein@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.3.1) Gecko/20030527 Debian/1.3.1-2
X-Accept-Language: en
MIME-Version: 1.0
To: =?UTF-8?B?SGVybWFubmkgSHl5dGnDpGzDpA==?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] PEG: Attacking against GISP
References: <1054892152.4089.58.camel@io.it.jyu.fi>
	<20030606110628.GA18385@tarzan.it.jyu.fi>
	<1054900410.4089.69.camel@io.it.jyu.fi>
In-Reply-To: <1054900410.4089.69.camel@io.it.jyu.fi>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
cc: Tuomas Lukka <lukka@mit.jyu.fi>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sat, 07 Jun 2003 18:28:37 -0000

Hermanni HyytiÃ¤lÃ¤ wrote:
> On Fri, 2003-06-06 at 14:06, Tuomas Lukka wrote:
> 
> 
>>>- A hostile (or faulty) peer(s) stores lot of dummy key-value-pairs with
>>>a certain key
>>
>>Does GISP allow several values per key?
> 
> Like DSHT ? No.

No, like a normal DHT, so that a lookup returns all values associated 
with a key. Yes, it does; we use it all the time.

- Benja




From MAILER-DAEMON Sat Jun 07 16:17:15 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Ok7m-0006cy-RF
	for mharc-gzz-dev@gnu.org; Sat, 07 Jun 2003 16:17:14 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Ok7l-0006Yf-5C
	for gzz-dev@nongnu.org; Sat, 07 Jun 2003 16:17:13 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Ok7j-0006Xp-Rv
	for gzz-dev@nongnu.org; Sat, 07 Jun 2003 16:17:12 -0400
Received: from mail.gmx.de ([213.165.64.20] helo=mail.gmx.net)
	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19Ok7j-0006Xl-En
	for gzz-dev@nongnu.org; Sat, 07 Jun 2003 16:17:11 -0400
Received: (qmail 13884 invoked by uid 65534); 7 Jun 2003 20:17:09 -0000
Received: from pD9E1270B.dip.t-dialin.net (EHLO gmx.de) (217.225.39.11)
  by mail.gmx.net (mp016) with SMTP; 07 Jun 2003 22:17:09 +0200
Message-ID: <3EE2481A.7010303@gmx.de>
Date: Sat, 07 Jun 2003 22:16:26 +0200
From: Benja Fallenstein <b.fallenstein@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.3.1) Gecko/20030527 Debian/1.3.1-2
X-Accept-Language: en
MIME-Version: 1.0
To: Tuomas Lukka <lukka@mit.jyu.fi>, GZZ developers <gzz-dev@nongnu.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Subject: [Gzz] Please fix shown resolution of imported files
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sat, 07 Jun 2003 20:17:13 -0000


Hi Tuomas,

tried importing a PDF file and it went fine, except that the resolution 
I ended up showing was very small (i.e. it looked clunky). Can you see 
whether it's the same for you and if so, fix it?

-b




From MAILER-DAEMON Sun Jun 08 05:58:33 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19OwwR-0006S8-FO
	for mharc-gzz-dev@gnu.org; Sun, 08 Jun 2003 05:58:23 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19OwwN-0006F4-59
	for gzz-dev@nongnu.org; Sun, 08 Jun 2003 05:58:19 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19OwwJ-00069G-Pk
	for gzz-dev@nongnu.org; Sun, 08 Jun 2003 05:58:16 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19OwwJ-00060t-0U
	for gzz-dev@nongnu.org; Sun, 08 Jun 2003 05:58:15 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h589w6aL026239;
	Sun, 8 Jun 2003 12:58:06 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h589w3rY006979;
	Sun, 8 Jun 2003 12:58:03 +0300 (EETDST)
Date: Sun, 8 Jun 2003 12:58:03 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
Message-ID: <20030608095803.GF21111@tarzan.it.jyu.fi>
References: <3EE2481A.7010303@gmx.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3EE2481A.7010303@gmx.de>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
cc: GZZ developers <gzz-dev@nongnu.org>
Subject: [Gzz] Re: Please fix shown resolution of imported files
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sun, 08 Jun 2003 09:58:21 -0000

On Sat, Jun 07, 2003 at 10:16:26PM +0200, Benja Fallenstein wrote:
> tried importing a PDF file and it went fine, except that the resolution 
> I ended up showing was very small (i.e. it looked clunky). Can you see 
> whether it's the same for you and if so, fix it?

Umm, could you check the resulting mipzip file with unzip -l and
post the result?

That will explain whether the fault is in creating the mipzip or simply
loading.

	Tuomas



From MAILER-DAEMON Sun Jun 08 07:33:31 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19OyPf-0007IJ-D4
	for mharc-gzz-dev@gnu.org; Sun, 08 Jun 2003 07:32:39 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19OyLZ-0005iH-SG
	for gzz-dev@nongnu.org; Sun, 08 Jun 2003 07:28:25 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19OyKf-0005Nc-C6
	for gzz-dev@nongnu.org; Sun, 08 Jun 2003 07:27:30 -0400
Received: from imap.gmx.net ([213.165.64.20] helo=mail.gmx.net)
	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19Oy44-000282-0D
	for gzz-dev@nongnu.org; Sun, 08 Jun 2003 07:10:20 -0400
Received: (qmail 30085 invoked by uid 65534); 8 Jun 2003 11:10:15 -0000
Received: from pD95880B4.dip.t-dialin.net (EHLO gmx.de) (217.88.128.180)
  by mail.gmx.net (mp027) with SMTP; 08 Jun 2003 13:10:15 +0200
Message-ID: <3EE31968.9030602@gmx.de>
Date: Sun, 08 Jun 2003 13:09:28 +0200
From: Benja Fallenstein <b.fallenstein@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.3.1) Gecko/20030527 Debian/1.3.1-2
X-Accept-Language: en
MIME-Version: 1.0
To: Tuomas Lukka <lukka@mit.jyu.fi>
Subject: Re: [Gzz] Re: Please fix shown resolution of imported files
References: <3EE2481A.7010303@gmx.de>
	<20030608095803.GF21111@tarzan.it.jyu.fi>
In-Reply-To: <20030608095803.GF21111@tarzan.it.jyu.fi>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
cc: GZZ developers <gzz-dev@nongnu.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sun, 08 Jun 2003 11:32:38 -0000

Tuomas Lukka wrote:
> On Sat, Jun 07, 2003 at 10:16:26PM +0200, Benja Fallenstein wrote:
> 
>>tried importing a PDF file and it went fine, except that the resolution 
>>I ended up showing was very small (i.e. it looked clunky). Can you see 
>>whether it's the same for you and if so, fix it?
> 
> 
> Umm, could you check the resulting mipzip file with unzip -l and
> post the result?
> 
> That will explain whether the fault is in creating the mipzip or simply
> loading.

I tried twice. Below the results.
-b

Archive: 
urn_-_x-storm_-_1.0_-_application__pdf,6lbquosafszyuhullwh3idsxkdxvbnte.26lrijjgukp4pfof7acylwnqlzugud56fgxaawi-160-1
   Length     Date   Time    Name
  --------    ----   ----    ----
         0  06-07-03 21:53   texformat
COMPRESSED_RGB_S3TC_DXT1_EXT
         0  06-07-03 21:53   origsize
0.6650390625x0.8603515625
   2097152  06-07-03 21:53   0
2048x2048
    524288  06-07-03 21:53   1
1024x1024
    131072  06-07-03 21:53   2
512x512
     32768  06-07-03 21:53   3
256x256
      8192  06-07-03 21:53   4
128x128
      2048  06-07-03 21:53   5
64x64
       512  06-07-03 21:53   6
32x32
       128  06-07-03 21:53   7
16x16
        32  06-07-03 21:53   8
8x8
         8  06-07-03 21:53   9
4x4
         8  06-07-03 21:53   10
2x2
         8  06-07-03 21:53   11
1x1
  --------                   -------
   2796216                   14 files



Archive: 
urn_-_x-storm_-_1.0_-_application__pdf,b73ieihyyne7l75cesxjk73kss36lo2v.ugucmvtxtktf6siaani4nbmid7odaaysy4uzkgy-160-1
   Length     Date   Time    Name
  --------    ----   ----    ----
         0  06-07-03 21:41   texformat
COMPRESSED_RGB_S3TC_DXT1_EXT
         0  06-07-03 21:41   origsize
0.625x0.83984375
   2097152  06-07-03 21:41   0
2048x2048
    524288  06-07-03 21:41   1
1024x1024
    131072  06-07-03 21:41   2
512x512
     32768  06-07-03 21:41   3
256x256
      8192  06-07-03 21:41   4
128x128
      2048  06-07-03 21:41   5
64x64
       512  06-07-03 21:41   6
32x32
       128  06-07-03 21:41   7
16x16
        32  06-07-03 21:41   8
8x8
         8  06-07-03 21:41   9
4x4
         8  06-07-03 21:41   10
2x2
         8  06-07-03 21:41   11
1x1
  --------                   -------
   2796216                   14 files




From MAILER-DAEMON Sun Jun 08 08:05:05 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Oyv3-00029P-4X
	for mharc-gzz-dev@gnu.org; Sun, 08 Jun 2003 08:05:05 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Oyv0-00028h-Pi
	for gzz-dev@nongnu.org; Sun, 08 Jun 2003 08:05:02 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Oyuz-00028S-Df
	for gzz-dev@nongnu.org; Sun, 08 Jun 2003 08:05:02 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Oysv-0001DJ-3q
	for gzz-dev@nongnu.org; Sun, 08 Jun 2003 08:02:53 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h58C2oaL000109;
	Sun, 8 Jun 2003 15:02:50 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h58C2m0O007433;
	Sun, 8 Jun 2003 15:02:48 +0300 (EETDST)
Date: Sun, 8 Jun 2003 15:02:48 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
Subject: Re: [Gzz] Re: Please fix shown resolution of imported files
Message-ID: <20030608120248.GG21111@tarzan.it.jyu.fi>
References: <3EE2481A.7010303@gmx.de>
	<20030608095803.GF21111@tarzan.it.jyu.fi> <3EE31968.9030602@gmx.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3EE31968.9030602@gmx.de>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
cc: GZZ developers <gzz-dev@nongnu.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sun, 08 Jun 2003 12:05:03 -0000

On Sun, Jun 08, 2003 at 01:09:28PM +0200, Benja Fallenstein wrote:
> Tuomas Lukka wrote:
> >On Sat, Jun 07, 2003 at 10:16:26PM +0200, Benja Fallenstein wrote:
> >
> >>tried importing a PDF file and it went fine, except that the resolution 
> >>I ended up showing was very small (i.e. it looked clunky). Can you see 
> >>whether it's the same for you and if so, fix it?
> >
> >
> >Umm, could you check the resulting mipzip file with unzip -l and
> >post the result?
> >
> >That will explain whether the fault is in creating the mipzip or simply
> >loading.
> 
> I tried twice. Below the results.
> -b
> 
> Archive: 
> urn_-_x-storm_-_1.0_-_application__pdf,6lbquosafszyuhullwh3idsxkdxvbnte.26lrijjgukp4pfof7acylwnqlzugud56fgxaawi-160-1
>   Length     Date   Time    Name
>  --------    ----   ----    ----
>         0  06-07-03 21:53   texformat
> COMPRESSED_RGB_S3TC_DXT1_EXT
>         0  06-07-03 21:53   origsize
> 0.6650390625x0.8603515625
>   2097152  06-07-03 21:53   0
> 2048x2048
>    524288  06-07-03 21:53   1

This shows that it has the largest mipmap level, so the problem
is a result of the desire to load the level not propagating properly.

Working on a document to explain in more detail.

	Tuomas




From MAILER-DAEMON Mon Jun 09 03:30:26 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PH6o-0006uk-JY
	for mharc-gzz-dev@gnu.org; Mon, 09 Jun 2003 03:30:26 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PH6l-0006tz-Pu
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 03:30:23 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PH6k-0006tn-FI
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 03:30:23 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19PH6j-0006tB-V2
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 03:30:22 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h597UBFb005932;	Mon, 9 Jun 2003 10:30:11 +0300
Subject: Re: [Gzz] PEG: Attacking against GISP
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
In-Reply-To: <3EE22E64.9020802@gmx.de>
References: <1054892152.4089.58.camel@io.it.jyu.fi>
	 <20030606110628.GA18385@tarzan.it.jyu.fi>
	 <1054900410.4089.69.camel@io.it.jyu.fi>  <3EE22E64.9020802@gmx.de>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1055143811.10663.10.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 09 Jun 2003 10:30:11 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti1.jyu.fi id
	h597UBFb005932
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
cc: Tuomas Lukka <lukka@mit.jyu.fi>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 09 Jun 2003 07:30:24 -0000

On Sat, 2003-06-07 at 21:26, Benja Fallenstein wrote:
> Hermanni Hyyti=E4l=E4 wrote:
> > On Fri, 2003-06-06 at 14:06, Tuomas Lukka wrote:
> >=20
> >=20
> >>>- A hostile (or faulty) peer(s) stores lot of dummy key-value-pairs =
with
> >>>a certain key
> >>
> >>Does GISP allow several values per key?
> >=20
> > Like DSHT ? No.
>=20
> No, like a normal DHT, so that a lookup returns all values associated=20
> with a key. Yes, it does; we use it all the time.
>=20


Sorry, I don't understand your answer completely :).

I understood Tuomas' question like "Does GISP support frequent fetches
and frequent *stores* of the same hash table key" (like DSHT supports),
not like "Only one value can be stored under a key at any given time in
regular DHTs. Thus, frequent retrievals but only one *store* is
supported. Does GISP support this ?"


-Hermanni








From MAILER-DAEMON Mon Jun 09 07:21:51 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PKiB-0007a3-Qa
	for mharc-gzz-dev@gnu.org; Mon, 09 Jun 2003 07:21:15 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PKfq-0006pB-MS
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 07:18:50 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PKeg-0006VK-9z
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 07:17:38 -0400
Received: from pop.gmx.net ([213.165.64.20] helo=mail.gmx.net)
	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19PKT4-0004t8-QC
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 07:05:39 -0400
Received: (qmail 2166 invoked by uid 65534); 9 Jun 2003 11:05:32 -0000
Received: from pD958849B.dip.t-dialin.net (EHLO gmx.de) (217.88.132.155)
  by mail.gmx.net (mp011) with SMTP; 09 Jun 2003 13:05:32 +0200
Message-ID: <3EE469CA.20306@gmx.de>
Date: Mon, 09 Jun 2003 13:04:42 +0200
From: Benja Fallenstein <b.fallenstein@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.3.1) Gecko/20030527 Debian/1.3.1-2
X-Accept-Language: en
MIME-Version: 1.0
To: =?ISO-8859-1?Q?Hermanni_Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] PEG: Attacking against GISP
References: <1054892152.4089.58.camel@io.it.jyu.fi>
	<20030606110628.GA18385@tarzan.it.jyu.fi>
	<1054900410.4089.69.camel@io.it.jyu.fi>  <3EE22E64.9020802@gmx.de>
	<1055143811.10663.10.camel@io.it.jyu.fi>
In-Reply-To: <1055143811.10663.10.camel@io.it.jyu.fi>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
cc: Tuomas Lukka <lukka@mit.jyu.fi>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 09 Jun 2003 11:21:14 -0000

Hermanni Hyytiälä wrote:
> On Sat, 2003-06-07 at 21:26, Benja Fallenstein wrote:
> 
>>Hermanni Hyytiälä wrote:
>>
>>>On Fri, 2003-06-06 at 14:06, Tuomas Lukka wrote:
>>>
>>>
>>>
>>>>>- A hostile (or faulty) peer(s) stores lot of dummy key-value-pairs with
>>>>>a certain key
>>>>
>>>>Does GISP allow several values per key?
>>>
>>>Like DSHT ? No.
>>
>>No, like a normal DHT, so that a lookup returns all values associated 
>>with a key. Yes, it does; we use it all the time.
> 
> Sorry, I don't understand your answer completely :).
> 
> I understood Tuomas' question like "Does GISP support frequent fetches
> and frequent *stores* of the same hash table key" (like DSHT supports),
> not like "Only one value can be stored under a key at any given time in
> regular DHTs. Thus, frequent retrievals but only one *store* is
> supported. Does GISP support this ?"

The point was simply whether different values can, at the same time, be 
associated with the same key. The answer is yes.

I'm sorry, but I don't really see the confusion here.

- Benja




From MAILER-DAEMON Mon Jun 09 07:28:54 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PKk4-0008SN-6R
	for mharc-gzz-dev@gnu.org; Mon, 09 Jun 2003 07:23:12 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PKhr-0007QR-K8
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 07:20:55 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PKdl-0006MC-4R
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 07:16:41 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19PKZe-0005Zu-8E
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 07:12:26 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h59BCNdc022952;	Mon, 9 Jun 2003 14:12:23 +0300
Subject: Re: [Gzz] PEG: Attacking against GISP
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
In-Reply-To: <3EE469CA.20306@gmx.de>
References: <1054892152.4089.58.camel@io.it.jyu.fi>
	 <20030606110628.GA18385@tarzan.it.jyu.fi>
	 <1054900410.4089.69.camel@io.it.jyu.fi>  <3EE22E64.9020802@gmx.de>
	 <1055143811.10663.10.camel@io.it.jyu.fi>  <3EE469CA.20306@gmx.de>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1055157143.10677.56.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 09 Jun 2003 14:12:23 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti2.jyu.fi id
	h59BCNdc022952
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
cc: Tuomas Lukka <lukka@mit.jyu.fi>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 09 Jun 2003 11:23:08 -0000

On Mon, 2003-06-09 at 14:04, Benja Fallenstein wrote:
> Hermanni Hyyti=E4l=E4 wrote:
> > On Sat, 2003-06-07 at 21:26, Benja Fallenstein wrote:
> >=20
> >>Hermanni Hyyti=E4l=E4 wrote:
> >>
> >>>On Fri, 2003-06-06 at 14:06, Tuomas Lukka wrote:
> >>>
> >>>
> >>>
> >>>>>- A hostile (or faulty) peer(s) stores lot of dummy key-value-pair=
s with
> >>>>>a certain key
> >>>>
> >>>>Does GISP allow several values per key?
> >>>
> >>>Like DSHT ? No.
> >>
> >>No, like a normal DHT, so that a lookup returns all values associated=
=20
> >>with a key. Yes, it does; we use it all the time.
> >=20
> > Sorry, I don't understand your answer completely :).
> >=20
> > I understood Tuomas' question like "Does GISP support frequent fetche=
s
> > and frequent *stores* of the same hash table key" (like DSHT supports=
),
> > not like "Only one value can be stored under a key at any given time =
in
> > regular DHTs. Thus, frequent retrievals but only one *store* is
> > supported. Does GISP support this ?"
>=20
> The point was simply whether different values can, at the same time, be=
=20
> associated with the same key. The answer is yes.
>=20
> I'm sorry, but I don't really see the confusion here.


I got confused by the way you *expressed* the matter, not the matter
itself ;).

So it's ok.



-Hermanni





From MAILER-DAEMON Mon Jun 09 08:12:34 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PLVq-0003Ws-3i
	for mharc-gzz-dev@gnu.org; Mon, 09 Jun 2003 08:12:34 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PLT6-0001rR-Ot
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 08:09:44 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PLSr-0001ZX-3v
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 08:09:31 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19PLSN-0000zp-1D
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 08:08:59 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h59C8tFb007221;	Mon, 9 Jun 2003 15:08:55 +0300
Subject: Second post: Re: [Gzz] PEG: Attacking against GISP
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
In-Reply-To: <3EE469CA.20306@gmx.de>
References: <1054892152.4089.58.camel@io.it.jyu.fi>
	 <20030606110628.GA18385@tarzan.it.jyu.fi>
	 <1054900410.4089.69.camel@io.it.jyu.fi>  <3EE22E64.9020802@gmx.de>
	 <1055143811.10663.10.camel@io.it.jyu.fi>  <3EE469CA.20306@gmx.de>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1055160535.10664.76.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 09 Jun 2003 15:08:55 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti1.jyu.fi id
	h59C8tFb007221
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
cc: Tuomas Lukka <lukka@mit.jyu.fi>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 09 Jun 2003 12:12:32 -0000

On Mon, 2003-06-09 at 14:04, Benja Fallenstein wrote:
> Hermanni Hyyti=E4l=E4 wrote:
> > On Sat, 2003-06-07 at 21:26, Benja Fallenstein wrote:
> >=20
> >>Hermanni Hyyti=E4l=E4 wrote:
> >>
> >>>On Fri, 2003-06-06 at 14:06, Tuomas Lukka wrote:
> >>>
> >>>
> >>>
> >>>>>- A hostile (or faulty) peer(s) stores lot of dummy key-value-pair=
s with
> >>>>>a certain key
> >>>>
> >>>>Does GISP allow several values per key?
> >>>
> >>>Like DSHT ? No.
> >>
> >>No, like a normal DHT, so that a lookup returns all values associated=
=20
> >>with a key. Yes, it does; we use it all the time.
> >=20
> > Sorry, I don't understand your answer completely :).
> >=20
> > I understood Tuomas' question like "Does GISP support frequent fetche=
s
> > and frequent *stores* of the same hash table key" (like DSHT supports=
),
> > not like "Only one value can be stored under a key at any given time =
in
> > regular DHTs. Thus, frequent retrievals but only one *store* is
> > supported. Does GISP support this ?"
>=20
> The point was simply whether different values can, at the same time, be=
=20
> associated with the same key. The answer is yes.
>=20


After a brief discussion with Tuomas, I decided to look GISP's source
code and found this:


First GISP.java

/**
 *  Applications' interface to GISP functionality.
 *  GISP, the Global Information Sharing Protocol, provides a     =20
distributed
 *  database mapping keys to values.
 *  This interface provides methods for querying the database and for
 *  inserting new mappings. An instance of this object represents a
 *  connection to one distributed database; if more than one database is
used,
 *  for example using different JXTA peer groups, <code>GISP</code>
objects
 *  need to be created for each of them.

...

*In each key/value pair, the key is a <code>String</code>. The value,
*on the other hand, can be either a <code>String</code> or an XML
*document,
*stored in a <code>byte[]</code> array. The simplest <code>insert</code>
*methods therefore are <code>insert(String keyword, String
*strdata)</code>
*and <code>insert(String keyword, byte[] xmldata)</code> for publishing
*new key/value pairs with <code>String</code> and XML values,
*respectively.
*Both <code>String</code> and XML values should be small: for example,
*rather than containing a file, they should contain information about
*which peer can be used to download a file.

...

/**
     * publish String data with keyword
     * with ttl
     *
     * @param key: keyword for the data
     * @param str: string data
     * @param ttl: milliseconds to live
     */
    public abstract void insert(String key, String str, long ttl);

- -

Then GISPXML.java (which is used, e.g., with the simulator and UDP-imp):

public void insert(String key, String str, long ttl){
	if(key =3D=3D null || str =3D=3D null || ttl <=3D 0){
	    throw new IllegalArgumentException("null argument(s)");
	}
	Element ele =3D DocumentHelper.createElement(ITEM);
	ele.addAttribute(KEYATTR, key);
	ele.addText(str);
	GInfo info =3D new GInfo(GHash.fromKey(key), ele,
			       getCurrentTime() + ttl, infoComparator);
	insertGInfo(info);
}

- -

GInfoManagerMemory.java:

...

private List datahouse =3D new ArrayList();

...

public void insertGInfo(GInfo newInfo){
	if(newInfo =3D=3D null){
	    LOG.error("insertGInfo: newInfo =3D=3D null");
	    return;
	}
	Element ele =3D (Element)newInfo.getData();
	if(ele.getDocument() =3D=3D null){
	    DocumentHelper.createDocument(ele);
	}
=09
	if(LOG.isDebugEnabled()){
	    LOG.debug("insertGInfo: element=3D" + ele.asXML());
	}

	synchronized(datahouse){
	    int index =3D datahouse.indexOf(newInfo);
	    if(index =3D=3D -1){
		datahouse.add(newInfo);
	    }else{
		GInfo info =3D (GInfo)datahouse.get(index);
		if(newInfo.isNewerThan(info)){
		    // update
		    info.update(newInfo);
		}
	    }
	}
}


- -
GInfo.java:

public synchronized void update(GInfo newInfo){
	assert newInfo !=3D null;
	assert this.equals(newInfo);

	expirationTime =3D newInfo.expirationTime;
	lastSentTable =3D newInfo.lastSentTable;
    }


- -

So it seems that GISP *replaces* the value for a
given key (updates time/date values) rather than appends the value
(e.g., to the end of a list) associated to a given key.

Given this, the answer is still no.



-Hermanni






From MAILER-DAEMON Mon Jun 09 10:45:02 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PNtK-00068l-UM
	for mharc-gzz-dev@gnu.org; Mon, 09 Jun 2003 10:44:58 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PNtG-0005xc-Bi
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 10:44:54 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PNtC-0005lj-07
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 10:44:53 -0400
Received: from mail.gmx.net ([213.165.64.20])
	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19PNt9-0005br-Hv
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 09 Jun 2003 10:44:47 -0400
Received: (qmail 27263 invoked by uid 65534); 9 Jun 2003 14:38:00 -0000
Received: from pD958849B.dip.t-dialin.net (EHLO gmx.de) (217.88.132.155)
  by mail.gmx.net (mp025) with SMTP; 09 Jun 2003 16:38:00 +0200
Message-ID: <3EE49B9A.3000705@gmx.de>
Date: Mon, 09 Jun 2003 16:37:14 +0200
From: Benja Fallenstein <b.fallenstein@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.3.1) Gecko/20030527 Debian/1.3.1-2
X-Accept-Language: en
MIME-Version: 1.0
To: =?ISO-8859-1?Q?Hermanni_Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: Second post: Re: [Gzz] PEG: Attacking against GISP
References: <1054892152.4089.58.camel@io.it.jyu.fi>
	<20030606110628.GA18385@tarzan.it.jyu.fi>
	<1054900410.4089.69.camel@io.it.jyu.fi>  <3EE22E64.9020802@gmx.de>
	<1055143811.10663.10.camel@io.it.jyu.fi>  <3EE469CA.20306@gmx.de>
	<1055160535.10664.76.camel@io.it.jyu.fi>
In-Reply-To: <1055160535.10664.76.camel@io.it.jyu.fi>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
cc: Tuomas Lukka <lukka@mit.jyu.fi>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 09 Jun 2003 14:44:57 -0000

Hermanni Hyytiälä wrote:
> On Mon, 2003-06-09 at 14:04, Benja Fallenstein wrote:
> 
>>Hermanni Hyytiälä wrote:
>>
>>>On Sat, 2003-06-07 at 21:26, Benja Fallenstein wrote:
>>>
>>>
>>>>Hermanni Hyytiälä wrote:
>>>>
>>>>
>>>>>On Fri, 2003-06-06 at 14:06, Tuomas Lukka wrote:
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>>>- A hostile (or faulty) peer(s) stores lot of dummy key-value-pairs with
>>>>>>>a certain key
>>>>>>
>>>>>>Does GISP allow several values per key?
>>>>>
>>>>>Like DSHT ? No.
>>>>
>>>>No, like a normal DHT, so that a lookup returns all values associated 
>>>>with a key. Yes, it does; we use it all the time.
>>>
>>>Sorry, I don't understand your answer completely :).
>>>
>>>I understood Tuomas' question like "Does GISP support frequent fetches
>>>and frequent *stores* of the same hash table key" (like DSHT supports),
>>>not like "Only one value can be stored under a key at any given time in
>>>regular DHTs. Thus, frequent retrievals but only one *store* is
>>>supported. Does GISP support this ?"
>>
>>The point was simply whether different values can, at the same time, be 
>>associated with the same key. The answer is yes.
>>
> 
> 
> 
> After a brief discussion with Tuomas, I decided to look GISP's source
> code and found this:

-- snip --

 > So it seems that GISP *replaces* the value for a
 > given key (updates time/date values) rather than appends the value
 > (e.g., to the end of a list) associated to a given key.
 >
 > Given this, the answer is still no.

So how can you explain away the fact that we have tests proving the 
contrary?

     org/nongnu/storm/modules/gispmap/GispP2PMap.test

which inherits from

     org/nongnu/storm/impl/p2p/P2PMap.meta

And the fact that Storm uses this and has worked fine so far, of course.

> Then GISPXML.java (which is used, e.g., with the simulator and UDP-imp):
> 
> public void insert(String key, String str, long ttl){
...
> 	GInfo info = new GInfo(GHash.fromKey(key), ele,
> 			       getCurrentTime() + ttl, infoComparator);
> 	insertGInfo(info);
> }
> 
> - -
> 
> GInfoManagerMemory.java:
> 
> public void insertGInfo(GInfo newInfo){
...
> 	    int index = datahouse.indexOf(newInfo);
> 	    if(index == -1){
> 		datahouse.add(newInfo);
> 	    }else{
> 		GInfo info = (GInfo)datahouse.get(index);
> 		if(newInfo.isNewerThan(info)){
> 		    // update
> 		    info.update(newInfo);
> 		}
> 	    }
...

The *intent* of the above might be to have only one value associated 
with one key; I'm not sure about that. However, I cannot see how this 
can be its effect, given the passing GISP tests and our use of GISP.

It is a bit hard to trace how GInfo.equals() works, but I think that it 
probably thinks two GInfos are equal only if the values as well as the 
keys in a key/value mapping are equal. I.e., the above will only update 
an entry if *both* the key and the value are equal.

Actually that sounds like a more sensible rationale: If both the key and 
the value are equal, it will update the timestamp on the key/value pair. 
However, I'm not sure this is correct either; I don't understand the 
source well enough. :-(

- Benja




From MAILER-DAEMON Tue Jun 10 03:34:41 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Pde6-0002ZN-LF
	for mharc-gzz-dev@gnu.org; Tue, 10 Jun 2003 03:34:18 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Pde0-0002QQ-C0
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 03:34:12 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Pddn-000286-2B
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 03:33:59 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Pddj-0001qR-0N
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 03:33:55 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5A7XjO8003866
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Tue, 10 Jun 2003 10:33:47 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5A7Xhjo015356
	for gzz-dev@mail.freesoftware.fsf.org;
	Tue, 10 Jun 2003 10:33:43 +0300 (EETDST)
Date: Tue, 10 Jun 2003 10:33:43 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Message-ID: <20030610073343.GH21111@tarzan.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Subject: [Gzz] Serializing URIs
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 10 Jun 2003 07:34:17 -0000

Benja (or someone), could you please put in comments to explain *what*
this code in Graphs is doing?

    public static Model toModel(ConstGraph g) {
        try {
            Model m = new ModelMem();
            for (Iterator i=g.findN_XAA_Iter(); i.hasNext();) {
                Object o = i.next();
                Resource sub = m.createResource(Nodes.toString(o));

                for (Iterator j=g.findN_1XA_Iter(o); j.hasNext();) {
                    Object p = j.next();

                    String s0 = Nodes.toString(p);
                    boolean hadAlpha = false;
                    int i0 = s0.length()-1;
                    Property prop;
                    while(true) {
                        char c0 = s0.charAt(i0);
                        if(Character.isLetter(c0)) hadAlpha = true;
                        else if(Character.isDigit(c0)) hadAlpha = false;
                        else if(hadAlpha) {
                            prop = m.createProperty(s0.substring(0, i0+1),
                                             s0.substring(i0+1));
                            break;
                        } else
                            throw new Error("Cannot serialize URI: "+s0);
                        i0--;
                        if(i0 < 0)
                            throw new Error("Cannot serialize URI: "+s0);
                    }



I'm getting a test failure but since I have absolutely not the faintest
of what this is supposed to be doing, I can't track it down.

	Tuomas




From MAILER-DAEMON Tue Jun 10 05:52:31 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Pfnr-0005Ig-5W
	for mharc-gzz-dev@gnu.org; Tue, 10 Jun 2003 05:52:31 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Pfno-0005An-7Z
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 05:52:28 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Pfnk-000542-NM
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 05:52:26 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Pfnj-000502-Jz
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 05:52:23 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5A9qFFb010921;	Tue, 10 Jun 2003 12:52:15 +0300
Subject: Re: Second post: Re: [Gzz] PEG: Attacking against GISP
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
In-Reply-To: <3EE49B9A.3000705@gmx.de>
References: <1054892152.4089.58.camel@io.it.jyu.fi>
	 <20030606110628.GA18385@tarzan.it.jyu.fi>
	 <1054900410.4089.69.camel@io.it.jyu.fi>  <3EE22E64.9020802@gmx.de>
	 <1055143811.10663.10.camel@io.it.jyu.fi>  <3EE469CA.20306@gmx.de>
	 <1055160535.10664.76.camel@io.it.jyu.fi>  <3EE49B9A.3000705@gmx.de>
Content-Type: text/plain
Organization: 
Message-Id: <1055238735.12787.15.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 10 Jun 2003 12:52:15 +0300
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
cc: Tuomas Lukka <lukka@mit.jyu.fi>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 10 Jun 2003 09:52:29 -0000

On Mon, 2003-06-09 at 17:37, Benja Fallenstein wrote:
> > 
> > public void insertGInfo(GInfo newInfo){
> ...
> > 	    int index = datahouse.indexOf(newInfo);
> > 	    if(index == -1){
> > 		datahouse.add(newInfo);
> > 	    }else{
> > 		GInfo info = (GInfo)datahouse.get(index);
> > 		if(newInfo.isNewerThan(info)){
> > 		    // update
> > 		    info.update(newInfo);
> > 		}
> > 	    }
> ...
> 
> The *intent* of the above might be to have only one value associated 
> with one key; I'm not sure about that. However, I cannot see how this 
> can be its effect, given the passing GISP tests and our use of GISP.
> 
> It is a bit hard to trace how GInfo.equals() works, but I think that it 
> probably thinks two GInfos are equal only if the values as well as the 
> keys in a key/value mapping are equal. I.e., the above will only update 
> an entry if *both* the key and the value are equal.

Today I (with the aid of Tuomas) looked GISP's source code more
carefully and noticed that GISP *does* support multiple values per key;
GInfo.equals() was implemented in a way that I didn't realize at first
how the method actually works with ArrayList.indexOf(): what matters
with the GInfo.equals() is that both the key and value for a given
object have to be equal when compared to "this" object's key and value
(as you cite above).

However, it's interesting to realize that there are no words in the DHT
literature about that DHT can be *open*, i.e., in the case a "hash
collision", hash table's single bucket stores multiple entries (like
Java's hash table).



-Hermanni





From MAILER-DAEMON Tue Jun 10 07:24:56 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PhFH-000619-RA
	for mharc-gzz-dev@gnu.org; Tue, 10 Jun 2003 07:24:55 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PhCe-00057I-CF
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 07:22:12 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PhB9-0004vA-Pw
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 07:20:42 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Pghr-0000iR-Q6
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 06:50:23 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5AAoIa6018098;
	Tue, 10 Jun 2003 13:50:18 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5AAoHJM021625;
	Tue, 10 Jun 2003 13:50:17 +0300 (EETDST)
Date: Tue, 10 Jun 2003 13:50:17 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Message-ID: <20030610105017.GI21111@tarzan.it.jyu.fi>
References: <E19PgSs-0004ZH-0o@subversions.gnu.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <E19PgSs-0004ZH-0o@subversions.gnu.org>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Subject: [Gzz] Re: [Gzz-commits] storm/doc/pegboard/attacking_gisp--hemppah
	peg.rst
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 10 Jun 2003 11:24:54 -0000

> There are two variations of this scenario. The first
> +variation is the static scenario in which peers do join or leave the system while
> +the test is running. In the second variation, peers can join and leave the
> +system while the test is running. 

You're missing a negation somewhere, I think.

> +We expect that GISP is able to route some queries to their destination peers eventually,
> +altough the performance of a lookup is expected to decrease. 

How do you suppose that will happen? 

You need to study the algorithms in GISP and explain how it will go.

Also, you need to estimate what you expect to find beforehand.

> Also, we except that
> +some of the queries are lost. With this test case, we wish to get more information
> +how big the lost rate is. 
> +
> +Simulation Process: 
> +- Create 90 normal peers in the network (ID is in the format "peer1 -peer90")
> +- Create 10 "dumb" peers in the network (ID is in the format "dumb1-dumb10")
> +- If necessary, the number of "dumb" peers can be changed (for more "clearer" 
> +  analysic etc)

This is not good - we want to understand
the scaling with both network size and dumb peer fraction.

> +- Create 100 data items in the network (the format is "key1-100", "value1-100")

Not good - several of the peers will have no items. I'd recommend enough
to have a good probability that each peer will have its own items.

> +- Perform 100 queries randomly (random peer selection (1-100) and random query 
> +  selection (1-100)) with *normal* peers

100 is rather little if the routing tables &c are expected to morph
during the experiment.

> +- Update all peers' routing information every loop pass 

How realistic is that? What changes will there be, if the network
itself is static?

	Tuomas




From MAILER-DAEMON Tue Jun 10 07:28:31 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PhId-0007DW-Fl
	for mharc-gzz-dev@gnu.org; Tue, 10 Jun 2003 07:28:23 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PhDq-0005NJ-Qn
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 07:23:26 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PhDo-0005MG-8V
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 07:23:25 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19PhDn-0005LZ-Dk
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 07:23:23 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5ABNJdc016705;	Tue, 10 Jun 2003 14:23:19 +0300
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030610105017.GI21111@tarzan.it.jyu.fi>
References: <E19PgSs-0004ZH-0o@subversions.gnu.org>
	 <20030610105017.GI21111@tarzan.it.jyu.fi>
Content-Type: text/plain
Organization: 
Message-Id: <1055244198.12788.46.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 10 Jun 2003 14:23:19 +0300
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Subject: [Gzz] 
 Re: [Gzz-commits] storm/doc/pegboard/attacking_gisp--hemppah
	peg.rst
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 10 Jun 2003 11:28:21 -0000


On Tue, 2003-06-10 at 13:50, Tuomas Lukka wrote:
> > There are two variations of this scenario. The first
> > +variation is the static scenario in which peers do join or leave the system while
> > +the test is running. In the second variation, peers can join and leave the
> > +system while the test is running. 
> 
> You're missing a negation somewhere, I think.


True (typo).

> 
> > +We expect that GISP is able to route some queries to their destination peers eventually,
> > +altough the performance of a lookup is expected to decrease. 
> 
> How do you suppose that will happen? 

Since GISP uses Chord-like routing table --> it is expected that the
destination can be found eventually using finger tables (i.e. "backup
links").


> 
> Also, you need to estimate what you expect to find beforehand.

Chord's fault tolerance properties which can be found from the original
paper ?

> 
> > Also, we except that
> > +some of the queries are lost. With this test case, we wish to get more information
> > +how big the lost rate is. 
> > +
> > +Simulation Process: 
> > +- Create 90 normal peers in the network (ID is in the format "peer1 -peer90")
> > +- Create 10 "dumb" peers in the network (ID is in the format "dumb1-dumb10")
> > +- If necessary, the number of "dumb" peers can be changed (for more "clearer" 
> > +  analysic etc)
> This is not good - we want to understand
> the scaling with both network size and dumb peer fraction.


Ok, I change it so that the number of normal peers and "dumb" peers can
increase/decrease.

> 
> > +- Create 100 data items in the network (the format is "key1-100", "value1-100")
> 
> Not good - several of the peers will have no items. I'd recommend enough
> to have a good probability that each peer will have its own items.

Is 5000 items enough ?

> 
> > +- Perform 100 queries randomly (random peer selection (1-100) and random query 
> > +  selection (1-100)) with *normal* peers
> 
> 100 is rather little if the routing tables &c are expected to morph
> during the experiment.


What about 2500 queries ?


> 
> > +- Update all peers' routing information every loop pass 
> 
> How realistic is that? What changes will there be, if the network
> itself is static?

For example, If a GISP peer is able determine that an another peer is
not "useful" because it doesn't  reply/forward queries.



-Hemppa




From MAILER-DAEMON Tue Jun 10 08:29:12 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PiEY-0008IW-5J
	for mharc-gzz-dev@gnu.org; Tue, 10 Jun 2003 08:28:14 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PiEP-00087u-S9
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 08:28:05 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PiEC-0007zY-R5
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 08:27:55 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19PiBU-00071R-CF
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 08:25:04 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5ACP0a6024164;
	Tue, 10 Jun 2003 15:25:00 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5ACOxbl023178;
	Tue, 10 Jun 2003 15:24:59 +0300 (EETDST)
Date: Tue, 10 Jun 2003 15:24:59 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Message-ID: <20030610122459.GJ21111@tarzan.it.jyu.fi>
References: <E19PgSs-0004ZH-0o@subversions.gnu.org>
	<20030610105017.GI21111@tarzan.it.jyu.fi>
	<1055244198.12788.46.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1055244198.12788.46.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5ACP0a6024164
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Subject: [Gzz] Re: [Gzz-commits] storm/doc/pegboard/attacking_gisp--hemppah
	peg.rst
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 10 Jun 2003 12:28:11 -0000

On Tue, Jun 10, 2003 at 02:23:19PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> > > +We expect that GISP is able to route some queries to their destina=
tion peers eventually,
> > > +altough the performance of a lookup is expected to decrease.=20
> >=20
> > How do you suppose that will happen?=20
>=20
> Since GISP uses Chord-like routing table --> it is expected that the
> destination can be found eventually using finger tables (i.e. "backup
> links").

Then, explain it in the PEG and give the estimates of the time.

> > Also, you need to estimate what you expect to find beforehand.
>=20
> Chord's fault tolerance properties which can be found from the original
> paper ?

State them here, too. Does the metric affect them?

You need to give here exactly the results you'd expect to see when
running the experiment.

> > > Also, we except that
> > > +some of the queries are lost. With this test case, we wish to get =
more information
> > > +how big the lost rate is.=20
> > > +
> > > +Simulation Process:=20
> > > +- Create 90 normal peers in the network (ID is in the format "peer=
1 -peer90")
> > > +- Create 10 "dumb" peers in the network (ID is in the format "dumb=
1-dumb10")
> > > +- If necessary, the number of "dumb" peers can be changed (for mor=
e "clearer"=20
> > > +  analysic etc)
> > This is not good - we want to understand
> > the scaling with both network size and dumb peer fraction.
>=20
> Ok, I change it so that the number of normal peers and "dumb" peers can
> increase/decrease.

No, you need to actually give the settings you plan to test.
And the scaling you expect.

> > > +- Create 100 data items in the network (the format is "key1-100", =
"value1-100")
> >=20
> > Not good - several of the peers will have no items. I'd recommend eno=
ugh
> > to have a good probability that each peer will have its own items.
>=20
> Is 5000 items enough ?

That's up to you to calculate ;)

> > > +- Perform 100 queries randomly (random peer selection (1-100) and =
random query=20
> > > +  selection (1-100)) with *normal* peers
> >=20
> > 100 is rather little if the routing tables &c are expected to morph
> > during the experiment.
>=20
> What about 2500 queries ?

Ditto.

> > > +- Update all peers' routing information every loop pass=20
> >=20
> > How realistic is that? What changes will there be, if the network
> > itself is static?
>=20
> For example, If a GISP peer is able determine that an another peer is
> not "useful" because it doesn't  reply/forward queries.

Is it able to do that?=20

Then you need to run until you've reached a steady state, and have an est=
imate
of how long that will take beforehand, and also about what the steady
state will be like.

	Tuomas



From MAILER-DAEMON Tue Jun 10 09:12:39 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PivW-0007SN-Mt
	for mharc-gzz-dev@gnu.org; Tue, 10 Jun 2003 09:12:38 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Pitj-0006up-Oe
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 09:10:47 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PiqP-00055Z-Ko
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 09:07:23 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Pipd-0004du-TP
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 09:06:34 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5AD6Wdc000850;	Tue, 10 Jun 2003 16:06:32 +0300
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030610122459.GJ21111@tarzan.it.jyu.fi>
References: <E19PgSs-0004ZH-0o@subversions.gnu.org>
	 <20030610105017.GI21111@tarzan.it.jyu.fi>
	 <1055244198.12788.46.camel@io.it.jyu.fi>
	 <20030610122459.GJ21111@tarzan.it.jyu.fi>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1055250392.12787.98.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 10 Jun 2003 16:06:32 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti2.jyu.fi id
	h5AD6Wdc000850
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Subject: [Gzz] 
 Re: [Gzz-commits] storm/doc/pegboard/attacking_gisp--hemppah
	peg.rst
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 10 Jun 2003 13:12:36 -0000

On Tue, 2003-06-10 at 15:24, Tuomas Lukka wrote:
> On Tue, Jun 10, 2003 at 02:23:19PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> > > > +We expect that GISP is able to route some queries to their desti=
nation peers eventually,
> > > > +altough the performance of a lookup is expected to decrease.=20
> > >=20
> > > How do you suppose that will happen?=20
> >=20
> > Since GISP uses Chord-like routing table --> it is expected that the
> > destination can be found eventually using finger tables (i.e. "backup
> > links").
>=20
> Then, explain it in the PEG and give the estimates of the time.

Should I include both the "big Os" and the number from the figures from
the Chord paper ?

>=20
> > > Also, you need to estimate what you expect to find beforehand.
> >=20
> > Chord's fault tolerance properties which can be found from the origin=
al
> > paper ?
>=20
> State them here, too. Does the metric affect them?

AFAIK, no. The routing table abstraction is the thing what matters
(e.g., how many connections a peer maintains in the routing table).


>=20
> > > > Also, we except that
> > > > +some of the queries are lost. With this test case, we wish to ge=
t more information
> > > > +how big the lost rate is.=20
> > > > +
> > > > +Simulation Process:=20
> > > > +- Create 90 normal peers in the network (ID is in the format "pe=
er1 -peer90")
> > > > +- Create 10 "dumb" peers in the network (ID is in the format "du=
mb1-dumb10")
> > > > +- If necessary, the number of "dumb" peers can be changed (for m=
ore "clearer"=20
> > > > +  analysic etc)
> > > This is not good - we want to understand
> > > the scaling with both network size and dumb peer fraction.
> >=20
> > Ok, I change it so that the number of normal peers and "dumb" peers c=
an
> > increase/decrease.
>=20
> No, you need to actually give the settings you plan to test.
> And the scaling you expect.

E.g.,=20

Setting 1: 90 normal peers, 10 "dumb" peers
Setting 2: 900 normal peers, 100 "dumb" peers
Setting 3: 9000 normal peers, 1000 "dumb" peers

Like this ?

>=20
> > > > +- Create 100 data items in the network (the format is "key1-100"=
, "value1-100")
> > >=20
> > > Not good - several of the peers will have no items. I'd recommend e=
nough
> > > to have a good probability that each peer will have its own items.
> >=20
> > Is 5000 items enough ?
>=20
> That's up to you to calculate ;)

In the original Chord paper, the authors use 100*N keys, if N is the
number of all peers. Perhaps I will use this setting.


>=20
> > > > +- Perform 100 queries randomly (random peer selection (1-100) an=
d random query=20
> > > > +  selection (1-100)) with *normal* peers
> > >=20
> > > 100 is rather little if the routing tables &c are expected to morph
> > > during the experiment.
> >=20
> > What about 2500 queries ?
>=20

Again, Chord paper cites, "Each peer in an experiment picked a random
set of keys to query from the system, and we measured the path length
required to resolve each query."

So, I'll use the same method...

>=20
> > > > +- Update all peers' routing information every loop pass=20
> > >=20
> > > How realistic is that? What changes will there be, if the network
> > > itself is static?
> >=20
> > For example, If a GISP peer is able determine that an another peer is
> > not "useful" because it doesn't  reply/forward queries.
>=20
> Is it able to do that?=20


Somewhat "yes", according to the GISP paper ("..a peer can discard
information of
unreachable peers"). However, I will look for the source how it's
implemented; is it able to recognize if a peer is alive "dumb" (answers
to PING) or not.


-Hermanni




From MAILER-DAEMON Tue Jun 10 14:29:51 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PnrY-00073e-Lc
	for mharc-gzz-dev@gnu.org; Tue, 10 Jun 2003 14:28:52 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PnrR-00070T-AY
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 14:28:45 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PnrP-0006xa-3f
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 14:28:44 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19PnrO-0006x0-JM
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 10 Jun 2003 14:28:42 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5AISda6015506;
	Tue, 10 Jun 2003 21:28:39 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5AISbk2026495;
	Tue, 10 Jun 2003 21:28:37 +0300 (EETDST)
Date: Tue, 10 Jun 2003 21:28:37 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Message-ID: <20030610182837.GL21111@tarzan.it.jyu.fi>
References: <E19PgSs-0004ZH-0o@subversions.gnu.org>
	<20030610105017.GI21111@tarzan.it.jyu.fi>
	<1055244198.12788.46.camel@io.it.jyu.fi>
	<20030610122459.GJ21111@tarzan.it.jyu.fi>
	<1055250392.12787.98.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1055250392.12787.98.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5AISda6015506
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Subject: [Gzz] Re: [Gzz-commits] storm/doc/pegboard/attacking_gisp--hemppah
	peg.rst
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 10 Jun 2003 18:28:46 -0000

On Tue, Jun 10, 2003 at 04:06:32PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> On Tue, 2003-06-10 at 15:24, Tuomas Lukka wrote:
> > On Tue, Jun 10, 2003 at 02:23:19PM +0300, Hermanni Hyyti=E4l=E4 wrote=
:
> > > > > +We expect that GISP is able to route some queries to their des=
tination peers eventually,
> > > > > +altough the performance of a lookup is expected to decrease.=20
> > > >=20
> > > > How do you suppose that will happen?=20
> > >=20
> > > Since GISP uses Chord-like routing table --> it is expected that th=
e
> > > destination can be found eventually using finger tables (i.e. "back=
up
> > > links").
> >=20
> > Then, explain it in the PEG and give the estimates of the time.
>=20
> Should I include both the "big Os" and the number from the figures from
> the Chord paper ?

Yes, *with* a reference.

> > > > > Also, we except that
> > > > > +some of the queries are lost. With this test case, we wish to =
get more information
> > > > > +how big the lost rate is.=20
> > > > > +
> > > > > +Simulation Process:=20
> > > > > +- Create 90 normal peers in the network (ID is in the format "=
peer1 -peer90")
> > > > > +- Create 10 "dumb" peers in the network (ID is in the format "=
dumb1-dumb10")
> > > > > +- If necessary, the number of "dumb" peers can be changed (for=
 more "clearer"=20
> > > > > +  analysic etc)
> > > > This is not good - we want to understand
> > > > the scaling with both network size and dumb peer fraction.
> > >=20
> > > Ok, I change it so that the number of normal peers and "dumb" peers=
 can
> > > increase/decrease.
> >=20
> > No, you need to actually give the settings you plan to test.
> > And the scaling you expect.
>=20
> E.g.,=20
>=20
> Setting 1: 90 normal peers, 10 "dumb" peers
> Setting 2: 900 normal peers, 100 "dumb" peers
> Setting 3: 9000 normal peers, 1000 "dumb" peers
>=20
> Like this ?

For example. Or just 9*10^n, 1*10^n for n =3D 1..3

And you also need to check different fractions.

> >=20
> > > > > +- Create 100 data items in the network (the format is "key1-10=
0", "value1-100")
> > > >=20
> > > > Not good - several of the peers will have no items. I'd recommend=
 enough
> > > > to have a good probability that each peer will have its own items.
> > >=20
> > > Is 5000 items enough ?
> >=20
> > That's up to you to calculate ;)
>=20
> In the original Chord paper, the authors use 100*N keys, if N is the
> number of all peers. Perhaps I will use this setting.

Sounds fairly reasonable; using too many is better than too few.

> > > > > +- Perform 100 queries randomly (random peer selection (1-100) =
and random query=20
> > > > > +  selection (1-100)) with *normal* peers
> > > >=20
> > > > 100 is rather little if the routing tables &c are expected to mor=
ph
> > > > during the experiment.
> > >=20
> > > What about 2500 queries ?
> >=20
>=20
> Again, Chord paper cites, "Each peer in an experiment picked a random
> set of keys to query from the system, and we measured the path length
> required to resolve each query."
>=20
> So, I'll use the same method...

Once or several times?

What is it that we want to measure? Among other things, whether gisp
is able to adapt itself...

> > > > > +- Update all peers' routing information every loop pass=20
> > > >=20
> > > > How realistic is that? What changes will there be, if the network
> > > > itself is static?
> > >=20
> > > For example, If a GISP peer is able determine that an another peer =
is
> > > not "useful" because it doesn't  reply/forward queries.
> >=20
> > Is it able to do that?=20
>=20
>=20
> Somewhat "yes", according to the GISP paper ("..a peer can discard
> information of
> unreachable peers"). However, I will look for the source how it's
> implemented; is it able to recognize if a peer is alive "dumb" (answers
> to PING) or not.

Indeed. This is pretty vital.

	Tuomas







From MAILER-DAEMON Wed Jun 11 03:18:02 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19PzrZ-0001Vx-0y
	for mharc-gzz-dev@gnu.org; Wed, 11 Jun 2003 03:17:41 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19PzrX-0001T9-EK
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 03:17:39 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19PzrT-0001KY-RF
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 03:17:37 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19PzrS-0001Hj-WA
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 03:17:35 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5B7HXdc024796;	Wed, 11 Jun 2003 10:17:33 +0300
Subject: Re: [Gzz] Re: [Gzz-commits]
	storm/doc/pegboard/attacking_gisp--hemppah peg.rst
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030610182837.GL21111@tarzan.it.jyu.fi>
References: <E19PgSs-0004ZH-0o@subversions.gnu.org>
	 <20030610105017.GI21111@tarzan.it.jyu.fi>
	 <1055244198.12788.46.camel@io.it.jyu.fi>
	 <20030610122459.GJ21111@tarzan.it.jyu.fi>
	 <1055250392.12787.98.camel@io.it.jyu.fi>
	 <20030610182837.GL21111@tarzan.it.jyu.fi>
Content-Type: text/plain
Organization: 
Message-Id: <1055315852.14924.11.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 11 Jun 2003 10:17:32 +0300
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 11 Jun 2003 07:17:40 -0000

On Tue, 2003-06-10 at 21:28, Tuomas Lukka wrote:

> > 
> > Again, Chord paper cites, "Each peer in an experiment picked a random
> > set of keys to query from the system, and we measured the path length
> > required to resolve each query."
> > 
> > So, I'll use the same method...
> 
> Once or several times?

The authors used in tests a network with n=2^k peers, where k varied
from 3 to 14.

So, several times (and the number of peers in the system varied).

Is this ok for us ?



-Hermanni




From MAILER-DAEMON Wed Jun 11 06:24:22 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Q2lm-00089h-5D
	for mharc-gzz-dev@gnu.org; Wed, 11 Jun 2003 06:23:54 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Q2lK-0007gs-Nv
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 06:23:26 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Q2kw-0007bB-SU
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 06:23:11 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Q2kF-0006nL-Mj
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 06:22:19 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5BAMD9D004132;
	Wed, 11 Jun 2003 13:22:13 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5BAMBrT006035;
	Wed, 11 Jun 2003 13:22:11 +0300 (EETDST)
Date: Wed, 11 Jun 2003 13:22:11 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] Re: [Gzz-commits]
	storm/doc/pegboard/attacking_gisp--hemppah peg.rst
Message-ID: <20030611102211.GN21111@tarzan.it.jyu.fi>
References: <E19PgSs-0004ZH-0o@subversions.gnu.org>
	<20030610105017.GI21111@tarzan.it.jyu.fi>
	<1055244198.12788.46.camel@io.it.jyu.fi>
	<20030610122459.GJ21111@tarzan.it.jyu.fi>
	<1055250392.12787.98.camel@io.it.jyu.fi>
	<20030610182837.GL21111@tarzan.it.jyu.fi>
	<1055315852.14924.11.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1055315852.14924.11.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5BAMD9D004132
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 11 Jun 2003 10:23:52 -0000

On Wed, Jun 11, 2003 at 10:17:32AM +0300, Hermanni Hyyti=E4l=E4 wrote:
> On Tue, 2003-06-10 at 21:28, Tuomas Lukka wrote:
>=20
> > >=20
> > > Again, Chord paper cites, "Each peer in an experiment picked a rand=
om
> > > set of keys to query from the system, and we measured the path leng=
th
> > > required to resolve each query."
> > >=20
> > > So, I'll use the same method...
> >=20
> > Once or several times?
>=20
> The authors used in tests a network with n=3D2^k peers, where k varied
> from 3 to 14.
>=20
> So, several times (and the number of peers in the system varied).
>=20
> Is this ok for us ?

I meant: does each peer try to get the set of keys it chose once or sever=
al times?

	Tuomas



From MAILER-DAEMON Wed Jun 11 06:44:35 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Q34l-00068J-IO
	for mharc-gzz-dev@gnu.org; Wed, 11 Jun 2003 06:43:31 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Q34K-0005yN-LR
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 06:43:04 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Q31d-0005Q4-LZ
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 06:40:19 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Q31D-0005GN-Vy
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 06:39:52 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5BAdnqO005511;	Wed, 11 Jun 2003 13:39:49 +0300
Subject: Re: [Gzz] Re: [Gzz-commits]
	storm/doc/pegboard/attacking_gisp--hemppah peg.rst
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030611102211.GN21111@tarzan.it.jyu.fi>
References: <E19PgSs-0004ZH-0o@subversions.gnu.org>
	 <20030610105017.GI21111@tarzan.it.jyu.fi>
	 <1055244198.12788.46.camel@io.it.jyu.fi>
	 <20030610122459.GJ21111@tarzan.it.jyu.fi>
	 <1055250392.12787.98.camel@io.it.jyu.fi>
	 <20030610182837.GL21111@tarzan.it.jyu.fi>
	 <1055315852.14924.11.camel@io.it.jyu.fi>
	 <20030611102211.GN21111@tarzan.it.jyu.fi>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1055327989.14925.22.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 11 Jun 2003 13:39:49 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti1.jyu.fi id
	h5BAdnqO005511
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 11 Jun 2003 10:43:29 -0000

On Wed, 2003-06-11 at 13:22, Tuomas Lukka wrote:
> On Wed, Jun 11, 2003 at 10:17:32AM +0300, Hermanni Hyyti=E4l=E4 wrote:
> > On Tue, 2003-06-10 at 21:28, Tuomas Lukka wrote:
> >=20
> > > >=20
> > > > Again, Chord paper cites, "Each peer in an experiment picked a ra=
ndom
> > > > set of keys to query from the system, and we measured the path le=
ngth
> > > > required to resolve each query."
> > > >=20
> > > > So, I'll use the same method...
> > >=20
> > > Once or several times?
> >=20
> > The authors used in tests a network with n=3D2^k peers, where k varie=
d
> > from 3 to 14.
> >=20
> > So, several times (and the number of peers in the system varied).
> >=20
> > Is this ok for us ?
>=20
> I meant: does each peer try to get the set of keys it chose once or sev=
eral times?
>=20
> 	Tuomas


The authors do not mention it. However, since the network they used in
the simulation was *static* (no "harmful" factors") and the purpose of
the simulation was just to measure lookup lengths we can assume that
each peer queried keys only once.

(OTOH, with the fault tolerant simulation (concurrent joins/leaves) they
do state that queries are not retried.)


-Hermanni




From MAILER-DAEMON Wed Jun 11 07:34:09 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Q3pk-0005zJ-42
	for mharc-gzz-dev@gnu.org; Wed, 11 Jun 2003 07:32:04 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Q3oi-0005PC-Ro
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 07:31:00 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Q3mh-0004u6-7M
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 07:28:58 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Q2zR-0004xF-VY
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 06:38:02 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5BAbx9D005095
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Wed, 11 Jun 2003 13:37:59 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5BAbw2u006231
	for gzz-dev@mail.freesoftware.fsf.org;
	Wed, 11 Jun 2003 13:37:58 +0300 (EETDST)
Date: Wed, 11 Jun 2003 13:37:58 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Message-ID: <20030611103758.GO21111@tarzan.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5BAbx9D005095
Subject: [Gzz] attacking gisp peg comments
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 11 Jun 2003 11:32:02 -0000

> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> PEG attacking_gisp--hemppah:
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> :Authors:  Hermanni Hyyti=E4l=E4
> :Date-Created: 2003-06-05
> :Last-Modified: $Date: 2003/06/11 09:28:52 $
> :Revision: $Revision: 1.21 $
> :Status:   Incomplete
>=20
> .. :Stakeholders:
> .. :Scope:    Major|Minor|Trivial|Cosmetic
> .. :Type:     META|Policy|Architecture|Interface|Implementation
>=20
> .. Affect-PEGs:
>=20
> This PEG document briefly describes the attack methods used by a "kille=
r"=20
> program. The program is intended to be used to test GISP_ P2P software'=
s=20
> robustness against hostile attacks. =20

Probably should explain "first version" and when you do the next one,
start a new peg.

The idea would be that you start the experiments once the PEG is accepted.

> In this document we mean with "hostile peer" as an entity which is able=
 to do a=20
> (limited/simplified/modified) number of regular GISP peer's functionali=
es
> in a way which may be harmful for the GISP network w.r.t. performance
> and redundancy. The harmfulness of a peer is a consequence of the fact
> that a peer is wilfully malicious.=20
>=20
> Disclaimer
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> This program is only used for research purposes and
> the goal is to improve GISP's resilience against hostile attacks.
>=20
> Additionally, the author of GISP has stated that he will address=20
> attacks as they become a problem. This inclines to think=20
> that writing an attack program will get the author to address=20
> that attacks.

inclines *us*

> We expect that GISP's fault tolerance is at least at the same level as =
Chord_'s=20
> fault tolerace since GISP's routing table is based on Chord's routing t=
able.
>=20
> Chord's general properties:
> - O(log^2 n) messages are required to join/leave operations=20
> - O(log n) lookup efficiency
> - Routing table maintains information about O(log n) peers
> - Routing table requires information about O(log n) of other peers
>   of *efficient* routing, but performance degrades gracefully
>   when that information is out of date=20
> - Only one piece of information per peer need to be corect in=20
>   order to guarantee correct (though slow) routing queries =20

Do you *KNOW* whether this works for GISP with dumb peers or not?


> Eventually, we will compare (and validate) our test results with the Ch=
ord's=20
> fault tolerance properties.

Not eventually.

YOU NEED TO MAKE A REAL INFORMATIVE STATEMENT ABOUT WHAT THE RESULTS
WILL BE THAT COULD BE RIGHT OR WRONG.

You're still shielding yourself by making really vague statements.
*NOT* good. You have to put yourself on the line by making some *real*
predictions which will be such that when *I* look at the predictions
and test results I can immediately see whether the predictions were right
or not -- and the probability of making the correct predictions randomly
should be really small!

> Simulation Process:=20
> - 9*10^k normal peers, 1*10^k "dumb" peers, where k =3D 1..3
> - n*10^k normal peers, d*10^k "dumb" peers, where k =3D 1..3, n =3D 1..=
9
>   and d =3D 1..9

What?

> - For a normal peers ID is in the format "peer1, peer2..." and for a "d=
umb" peer=20
>   ID is in the format "dumb1, dumb2..."=20

Is this relevant? Just say "different spaces"

> - Create 5000 key/value items in the network (the format is "key1, key2=
...",=20
>   "value1, value2...")

Same here: these are irrelevant details.

> - Perform 2500 queries randomly with *normal* peers (random peer select=
ion =20
>   and random query selection)

Why 2500?

> - Test case is performed with loop (e.g., while(true) etc)

Another irrelevant point

> - "Distributed" peermap is based on Java's HashMap data structure (sync=
hronized)
> - Use org.apache.log4j package for logging information

And some more.


> Scenario #2 (dynamic "dumb"):=20

Let's postpone this to the next PEG.

> Issues
> =3D=3D=3D=3D=3D=3D
>=20
> ISSUE: When should we discuss this with the author of GISP?
>=20
> ISSUE: Does GISP support "peer-choice" during lookups?
>=20
> ISSUE: Does GISP peer is able determine if an another peer is
>        not "useful" or not (not just PING scenario)  ("a peer can disca=
rd
>        information of unreachable peers")   =20

"Does .. is able" isn't proper English.

	Tuomas



From MAILER-DAEMON Wed Jun 11 08:11:33 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Q4RH-000895-64
	for mharc-gzz-dev@gnu.org; Wed, 11 Jun 2003 08:10:51 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Q4Qr-0007wx-Ku
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 08:10:25 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Q4QY-0007oo-Gw
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 08:10:08 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Q4P1-0007Um-Mw
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 08:08:31 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5BC8MqO028660;	Wed, 11 Jun 2003 15:08:22 +0300
Subject: Re: [Gzz] attacking gisp peg comments
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030611103758.GO21111@tarzan.it.jyu.fi>
References: <20030611103758.GO21111@tarzan.it.jyu.fi>
Content-Type: text/plain
Organization: 
Message-Id: <1055333301.14905.61.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 11 Jun 2003 15:08:21 +0300
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 11 Jun 2003 12:10:37 -0000


On Wed, 2003-06-11 at 13:37, Tuomas Lukka wrote:

> > We expect that GISP's fault tolerance is at least at the same level as Chord_'s 
> > fault tolerace since GISP's routing table is based on Chord's routing table.
> > 
> > Chord's general properties:
> > - O(log^2 n) messages are required to join/leave operations 
> > - O(log n) lookup efficiency
> > - Routing table maintains information about O(log n) peers
> > - Routing table requires information about O(log n) of other peers
> >   of *efficient* routing, but performance degrades gracefully
> >   when that information is out of date 
> > - Only one piece of information per peer need to be corect in 
> >   order to guarantee correct (though slow) routing queries  
> 
> Do you *KNOW* whether this works for GISP with dumb peers or not?

No I don't. The problem is that we don't know how similar GISP's routing
table *really* (and overlay network) is to Chord's one.

That's why I sent an email to Daishi earlier today. 


> 
> 
> > Eventually, we will compare (and validate) our test results with the Chord's 
> > fault tolerance properties.
> 
> Not eventually.
> 
> YOU NEED TO MAKE A REAL INFORMATIVE STATEMENT ABOUT WHAT THE RESULTS
> WILL BE THAT COULD BE RIGHT OR WRONG.

This PEG (and this part of the document) is not ready yet... 

OTOH, is that obvious that the "RIGHT" results are (almost) similar
to Chord's results which are listed in the PEG already, e.g., 
"Result: 20% of lookups fail, when 20% of peers are failed" ?

Hm...so I should *emphasize* (i.e., repeat) that we expect similar
results as Chord has gained in the paper, or...?


> > Simulation Process: 
> > - 9*10^k normal peers, 1*10^k "dumb" peers, where k = 1..3
> > - n*10^k normal peers, d*10^k "dumb" peers, where k = 1..3, n = 1..9
> >   and d = 1..9
> 
> What?

Yes, indistinctly expressed. I already tried to make this better.


> > - Perform 2500 queries randomly with *normal* peers (random peer selection  
> >   and random query selection)
> 
> Why 2500?


Uh, I forgot to update this to 100*N (which I talked about yesterday).
Will do it.



-Hermanni




From MAILER-DAEMON Wed Jun 11 08:43:56 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Q4xF-0006xX-2O
	for mharc-gzz-dev@gnu.org; Wed, 11 Jun 2003 08:43:53 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Q4v8-0005k5-RN
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 08:41:42 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Q4uj-0005Vj-Af
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 08:41:18 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Q4on-0003zV-0d
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 08:35:09 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5BCZ6jX012300;
	Wed, 11 Jun 2003 15:35:06 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5BCZ467009217;
	Wed, 11 Jun 2003 15:35:04 +0300 (EETDST)
Date: Wed, 11 Jun 2003 15:35:04 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] attacking gisp peg comments
Message-ID: <20030611123504.GP21111@tarzan.it.jyu.fi>
References: <20030611103758.GO21111@tarzan.it.jyu.fi>
	<1055333301.14905.61.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1055333301.14905.61.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5BCZ6jX012300
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 11 Jun 2003 12:43:45 -0000

On Wed, Jun 11, 2003 at 03:08:21PM +0300, Hermanni Hyyti=E4l=E4 wrote:
>=20
> On Wed, 2003-06-11 at 13:37, Tuomas Lukka wrote:
>=20
> > > We expect that GISP's fault tolerance is at least at the same level=
 as Chord_'s=20
> > > fault tolerace since GISP's routing table is based on Chord's routi=
ng table.
> > >=20
> > > Chord's general properties:
> > > - O(log^2 n) messages are required to join/leave operations=20
> > > - O(log n) lookup efficiency
> > > - Routing table maintains information about O(log n) peers
> > > - Routing table requires information about O(log n) of other peers
> > >   of *efficient* routing, but performance degrades gracefully
> > >   when that information is out of date=20
> > > - Only one piece of information per peer need to be corect in=20
> > >   order to guarantee correct (though slow) routing queries =20
> >=20
> > Do you *KNOW* whether this works for GISP with dumb peers or not?
>=20
> No I don't. The problem is that we don't know how similar GISP's routin=
g
> table *really* (and overlay network) is to Chord's one.
>=20
> That's why I sent an email to Daishi earlier today.=20

Umm, I think you send too many emails without first trying to find
out yourself (as in the DHT/DSHT thing). Emailing the author
should not be your *FIRST* option, before you even understand the problem
well. =20

Personally, I like it when people ask me things but I don't like
it when unknown people ask me uninformed questions. Both "un"s are needed=
:
it's ok when people I know ask me uninformed questions - then I'll do my =
best
to make them informed. Also, when unknown people ask me informed question=
s
it's a delight to answer. But the unknown-uninformed combination is not n=
ice -
it's something that easily leads to "death of the internet", since LOTS
of unknown people could ask the author of one piece of software uninforme=
d=20
questions, which they should have asked from someone near to them.

The source is there and is fairly small - why don't you read and understa=
nd
it instead?

Then you could make informed answers instead of just echoing what other p=
eople
say even which is especially harmful if they were talking about completel=
y=20
different concepts using the same names.

> > > Eventually, we will compare (and validate) our test results with th=
e Chord's=20
> > > fault tolerance properties.
> >=20
> > Not eventually.
> >=20
> > YOU NEED TO MAKE A REAL INFORMATIVE STATEMENT ABOUT WHAT THE RESULTS
> > WILL BE THAT COULD BE RIGHT OR WRONG.
>=20
> This PEG (and this part of the document) is not ready yet...=20
>=20
> OTOH, is that obvious that the "RIGHT" results are (almost) similar
> to Chord's results which are listed in the PEG already, e.g.,=20
> "Result: 20% of lookups fail, when 20% of peers are failed" ?
>=20
> Hm...so I should *emphasize* (i.e., repeat) that we expect similar
> results as Chord has gained in the paper, or...?

No, you should not say "expect similar" - it's far too vague.
You should make explicit statements as to what the results will
be.

Stating

	If fraction f of the peers are dumb, then fraction f of the lookups
	will fail at all network sizes.

would be ok. This is an accurate, measurable statement which we can then
start seeing whether it's right or wrong.

Whether it *is* right or wrong, depends of course on the exact behaviour
of GISP which it's your job to find out about (and not emailing the autho=
r
except as last recourse).

	Tuomas




From MAILER-DAEMON Wed Jun 11 10:12:18 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Q6KW-0004E7-Ho
	for mharc-gzz-dev@gnu.org; Wed, 11 Jun 2003 10:12:00 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Q6KL-0003zm-LP
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 10:11:49 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Q6K9-0003bF-VD
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 10:11:39 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Q6Jn-00037F-Vh
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 11 Jun 2003 10:11:16 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5BEBDjX021079
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Wed, 11 Jun 2003 17:11:13 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5BEBC11011576
	for gzz-dev@mail.freesoftware.fsf.org;
	Wed, 11 Jun 2003 17:11:12 +0300 (EETDST)
Date: Wed, 11 Jun 2003 17:11:12 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Message-ID: <20030611141112.GR21111@tarzan.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Subject: [Gzz] Coming to jkl tomorrow
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 11 Jun 2003 14:11:59 -0000

Just a note: I'll be in jkl on thursday.

	Tuomas



From MAILER-DAEMON Fri Jun 13 06:29:31 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19QlnI-0003er-VQ
	for mharc-gzz-dev@gnu.org; Fri, 13 Jun 2003 06:28:28 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Qln2-0003Sn-QM
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 13 Jun 2003 06:28:12 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19QlkZ-0002YT-1E
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 13 Jun 2003 06:25:41 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19QlkU-0002MU-Fn
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 13 Jun 2003 06:25:34 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5DAPPjX004111
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Fri, 13 Jun 2003 13:25:25 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5DAPPKP005744
	for gzz-dev@mail.freesoftware.fsf.org;
	Fri, 13 Jun 2003 13:25:25 +0300 (EETDST)
Date: Fri, 13 Jun 2003 13:25:25 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Message-ID: <20030613102525.GS21111@tarzan.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Subject: [Gzz] [lac@ecs.soton.ac.uk: ACM HT03: Short Paper Results]
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 13 Jun 2003 10:28:20 -0000

Short papers also 25%...

	Tuomas

----- Forwarded message from Les Carr <lac@ecs.soton.ac.uk> -----

Return-Path: lac@ecs.soton.ac.uk
Received: from jokotaas.iki.fi (jokotaas.iki.fi [212.16.100.2])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5D6f1jX017133
	for <lukka@jane.math.jyu.fi>; Fri, 13 Jun 2003 09:41:03 +0300 (EETDST)
Received: from raven.ecs.soton.ac.uk (raven.ecs.soton.ac.uk [152.78.70.1])
	by jokotaas.iki.fi (8.11.6p2/8.11.6) with ESMTP id h5D6f1O23174
	for <lukka@iki.fi>; Fri, 13 Jun 2003 09:41:01 +0300 (EEST)
Received: from pigeon.ecs.soton.ac.uk (ns1 [152.78.68.1])
	by raven.ecs.soton.ac.uk (8.9.3/8.9.3) with ESMTP id HAA10373;
	Fri, 13 Jun 2003 07:40:57 +0100 (BST)
Received: from login.ecs.soton.ac.uk (IDENT:root@login [152.78.68.162])
	by pigeon.ecs.soton.ac.uk (8.9.3p2/8.9.3) with ESMTP id HAA09084;
	Fri, 13 Jun 2003 07:40:54 +0100 (BST)
From: Les Carr <lac@ecs.soton.ac.uk>
Received: (from lac@localhost)
	by login.ecs.soton.ac.uk (8.11.6/8.11.6) id h5D6eqP31561;
	Fri, 13 Jun 2003 07:40:52 +0100
Date: Fri, 13 Jun 2003 07:40:52 +0100
Message-Id: <200306130640.h5D6eqP31561@login.ecs.soton.ac.uk>
To: lukka@iki.fi
Subject: ACM HT03: Short Paper Results
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi

I am sorry to inform you that your submission
(paper 174, Bridging  Javadoc and design documentation via UML diagram image maps) 
has not been chosen as a short paper for the ACM Hypertext 2003 conference.
The quality has remained high with 25% of submissions being accepted,
and despite the short timescale, most submissions were reviewed by
four referees.

Thank you for your interest in HT03. We are anticipating a lively conference
and would value your participation.
---
Les Carr & Lynda Hardman
HT03 PC CoChairs

===================== REVIEWS ====================

Score: 4
	 Whilst the approach sounds interesting, many of the details
	 that would really be useful have not been included in the
	 paper. For example, the process for indentifying and
	 managing the document links is largely ignored. It is also
	 unclear whether the approach for constructing the diagrams
	 etc. is predominantly manual (in which case it would seem
	 prohibitively expensive except in trivial cass where it
	 is not needed). There is also no information on an evaluation
	 of the usefulness to users etc.

Score: 5
	 Sort of a nifty project in some senses, but on te other
	 hand, hard to see if this is really a research contribution
	 or not...

Score: 7


Score: 4
	 First, let me say that I really enjoyed reading this paper.
	 I liked your style.  I found the illustrations refreshing.
	 I found your motivation in the introduction right on the
	 mark --- and: I would love to have this solution for our
	 developments.  All said, I had many reasons why I would
	 have liked to accept this paper. The problem is that the
	 entire paper, as written, should serve as the introduction
	 to the real paper that you should have written. The HT
	 paper that I would have accepted would have devoted 1 page
	 describing your approach and one page providing *results*
	 - showing that this approach really improves the development.
	 This is the real contribution of your research.  You spend
	 great detail on the tool and its relevance for software
	 development.  But your domain is Software Engineering and
	 Hypertext. You describe how to navigate to the relevant
	 UML-designs, but almost nothing addresses its impact on
	 the development. I am pretty sure that the implementation
	 was quite a lot of work, but

----- End forwarded message -----



From MAILER-DAEMON Fri Jun 13 07:57:54 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19QnBM-0004R4-B5
	for mharc-gzz-dev@gnu.org; Fri, 13 Jun 2003 07:57:24 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Qn7r-0003aK-Fv
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 13 Jun 2003 07:53:47 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Qn7Z-00034r-9M
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 13 Jun 2003 07:53:30 -0400
Received: from verso.st.jyu.fi ([130.234.4.66])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Qn7L-0002h3-6V
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 13 Jun 2003 07:53:15 -0400
Received: from verso.st.jyu.fi (localhost.localdomain [127.0.0.1])
	by verso.st.jyu.fi (8.12.8/8.12.8) with ESMTP id h5DBr6ND002070
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Fri, 13 Jun 2003 14:53:06 +0300
Received: from localhost (atsoukka@localhost)h5DBr6Go002066
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Fri, 13 Jun 2003 14:53:06 +0300
X-Authentication-Warning: verso.st.jyu.fi: atsoukka owned process doing -bs
Date: Fri, 13 Jun 2003 14:53:06 +0300 (EEST)
From: Asko Soukka <asko.soukka@iki.fi>
X-X-Sender: atsoukka@verso.st.jyu.fi
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Message-ID: <Pine.LNX.4.44.0306131436280.582-100000@verso.st.jyu.fi>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Gzz] Navidoc make targets
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
Reply-To: Asko Soukka <asko.soukka@iki.fi>
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 13 Jun 2003 11:57:22 -0000

Hei,

I've been adding several Navidoc-related targets in all subprojects' 
makefiles (storm, alph, callg, glmosaictext, fenfire, storm, loom, 
navidoc). Updating same targets in eight makefiles can be exhausting, so 
I hope I haven't make much typos :) Please, report me about possibly 
broken targets.

As Tuomas noticed, there wasn't target for converting reST into HTML and
linking UML diagrams :/

Here is a list with short description:

html		; converts given $RST into html
latex		; converts given $RST into latex
html-loop	; loops converting given $RST into html
latex-loop	; loops converting given $RST into latex
naviloop	; loops converting given $RST into html, creating imagemaps
		  for diagrams and embedding diagrams into their
                  elements' target files
peg		; asks a few questions and creates directory, .cvsignore
                  and a template for a new peg

The following targets are for project's whole documentation:

navidoc		; compiles reST under project/doc(s)/
navilink	; imagemaps diagrams under project/doc(s)/

Navidoc itself has also:

imagemap	; create imagemaps for diagrams in $HTML, embed diagrams 
		  into their elements' target files

$RST can be any single file or a directory, when all files with .rst 
suffix will be handled.

$HTML can be any single HTML file or a directory, when all files with 
.html suffix will be handled.

-- 
Asko Soukka <asko.soukka@iki.fi>
<http://www.iki.fi/asko.soukka/>




From MAILER-DAEMON Fri Jun 13 16:33:38 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19QvEk-0006SV-QR
	for mharc-gzz-dev@gnu.org; Fri, 13 Jun 2003 16:33:26 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19QvEV-00061I-Ge
	for gzz-dev@nongnu.org; Fri, 13 Jun 2003 16:33:11 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19QvED-0005Gi-T5
	for gzz-dev@nongnu.org; Fri, 13 Jun 2003 16:32:56 -0400
Received: from imap.gmx.net ([213.165.64.20] helo=mail.gmx.net)
	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19QvCz-0004nX-FG
	for gzz-dev@nongnu.org; Fri, 13 Jun 2003 16:31:37 -0400
Received: (qmail 5203 invoked by uid 65534); 13 Jun 2003 20:31:26 -0000
Received: from pD9E12373.dip.t-dialin.net (EHLO gmx.de) (217.225.35.115)
  by mail.gmx.net (mp026) with SMTP; 13 Jun 2003 22:31:26 +0200
Message-ID: <3EEA344A.4060003@gmx.de>
Date: Fri, 13 Jun 2003 22:30:02 +0200
From: Benja Fallenstein <b.fallenstein@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.3.1) Gecko/20030527 Debian/1.3.1-2
X-Accept-Language: en
MIME-Version: 1.0
To: GZZ developers <gzz-dev@nongnu.org>
Content-Type: multipart/mixed;
 boundary="------------000300080100070906070800"
Subject: [Gzz] [Fwd: ACM HT03: Short Paper Results]
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 13 Jun 2003 20:33:24 -0000

This is a multi-part message in MIME format.
--------------000300080100070906070800
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit


--------------000300080100070906070800
Content-Type: message/rfc822;
 name="ACM HT03: Short Paper Results"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="ACM HT03: Short Paper Results"

Return-Path: <lac@ecs.soton.ac.uk>
X-Flags: 0000
Delivered-To: GMX delivery to b.fallenstein@gmx.de
Received: (qmail 10129 invoked by uid 65534); 13 Jun 2003 06:41:09 -0000
Received: from raven.ecs.soton.ac.uk (EHLO raven.ecs.soton.ac.uk)
	(152.78.70.1)
	by mx0.gmx.net (mx011) with SMTP; 13 Jun 2003 08:41:09 +0200
Received: from pigeon.ecs.soton.ac.uk (ns1 [152.78.68.1])
	by raven.ecs.soton.ac.uk (8.9.3/8.9.3) with ESMTP id HAA10421;
	Fri, 13 Jun 2003 07:41:00 +0100 (BST)
Received: from login.ecs.soton.ac.uk (IDENT:root@login [152.78.68.162])
	by pigeon.ecs.soton.ac.uk (8.9.3p2/8.9.3) with ESMTP id HAA09127;
	Fri, 13 Jun 2003 07:40:57 +0100 (BST)
From: Les Carr <lac@ecs.soton.ac.uk>
Received: (from lac@localhost)
	by login.ecs.soton.ac.uk (8.11.6/8.11.6) id h5D6eqk31613;
	Fri, 13 Jun 2003 07:40:52 +0100
Date: Fri, 13 Jun 2003 07:40:52 +0100
Message-Id: <200306130640.h5D6eqk31613@login.ecs.soton.ac.uk>
To: b.fallenstein@gmx.de
Subject: ACM HT03: Short Paper Results
X-GMX-Antivirus: -1 (not scanned, may not use virus scanner)
X-GMX-Antispam: 0 (Mail was not recognized as spam)

I am pleased to inform you that your submission
(paper 192, Storm: Using P2P to make the desktop part of the Web) 
has been accepted as a short paper for the ACM Hypertext 2003 conference.
The quality has remained high with 25% of submissions being accepted,
and despite the short timescale, most submissions were reviewed by
four referees.

The timescale for the final version of your paper is very tight - we
need to receive the final, camera-ready copy from you by Friday 20th June.
When preparing the final version, you should take into account the referees'
comments (below) and the formatting instructions on
http://www.ht03.org/submissions.html

Thank you for your interest in HT03. We are anticipating a lively conference
and look forward to your participation.
---
Les Carr & Lynda Hardman
HT03 PC CoChairs

===================== REVIEWS ====================

Score: 8
	 An interesting look at a significant work in progress

Score: 7
	 Nice paper, look forward to seeing it in print. But before
	 it does, please don`t use the name "Storm" so many times
	 -sounds as if you are trying to sell us a product!  Also,
	 you have completely misunderstood what Schneier says about
	 cryptographic hashes in your reference to [10] - it is
	 *NOT* "practically impossible" that two documents will
	 _have_ the same hash (in fact, it is absolutely certainly
	 going to  happen), but rather, it is practically impossible
	 to _find_ any two documents that have the same hash. This
	 is the "collision resistance" property and it means that
	 is is hard to discover other documents with the same hash,
	 but it is quite easy for such to exist, purely on the
	 pigeon-hole principle (whether alternative documents with
	 the same hash are meaningful is another matter).

Score: 5
	 The motivation of `including the desktop` is valid, although
	 the UI changes that your suggested approach would require
	 are too numerous for such a change to be usable - a fact
	 to which you allude in Section 3.   It is also unclear
	 how using a straightforward hashing function, such as
	 SHA-1 or MD5, would suffice to cater for all resources in
	 the application space without collisions (I would argue
	 that it is insufficient).   Further, the choice of URN
	 naming is not new (e.g. eDonkey`s URN format for file
	 sharing), and there is no discussion of the impact of
	 moving from a structured naming scheme (of which users
	 may rightly or wrongly make assumptions regarding content)
	 to an opaque one.  Discussion of Storm`s security model
	 would help the reader assess the utility of having `all
	 documents stored on their local hard disk` included
	 automagically by the Storm system, especially in light of
	 recent discoveries of other p2p filesharing tools and
	 their somewhat questionable behaviour of file publication.
	 Th



--------------000300080100070906070800--




From MAILER-DAEMON Fri Jun 13 16:42:39 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19QvNf-0004iQ-7X
	for mharc-gzz-dev@gnu.org; Fri, 13 Jun 2003 16:42:39 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19QvNd-0004fP-RY
	for gzz-dev@nongnu.org; Fri, 13 Jun 2003 16:42:37 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19QvNc-0004eM-20
	for gzz-dev@nongnu.org; Fri, 13 Jun 2003 16:42:37 -0400
Received: from pop.gmx.de ([213.165.64.20] helo=mail.gmx.net)
	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19QvNb-0004e0-Im
	for gzz-dev@nongnu.org; Fri, 13 Jun 2003 16:42:35 -0400
Received: (qmail 5988 invoked by uid 65534); 13 Jun 2003 20:42:33 -0000
Received: from pD9E12373.dip.t-dialin.net (EHLO gmx.de) (217.225.35.115)
  by mail.gmx.net (mp004) with SMTP; 13 Jun 2003 22:42:33 +0200
Message-ID: <3EEA370B.7030605@gmx.de>
Date: Fri, 13 Jun 2003 22:41:47 +0200
From: Benja Fallenstein <b.fallenstein@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.3.1) Gecko/20030527 Debian/1.3.1-2
X-Accept-Language: en
MIME-Version: 1.0
To: Les Carr <lac@ecs.soton.ac.uk>
References: <200306130640.h5D6eqk31613@login.ecs.soton.ac.uk>
In-Reply-To: <200306130640.h5D6eqk31613@login.ecs.soton.ac.uk>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
cc: GZZ developers <gzz-dev@nongnu.org>
Subject: [Gzz] Re: ACM HT03: Short Paper Results
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 13 Jun 2003 20:42:38 -0000


Dear Les,

thank you for your email! It looks as if there is something missing from 
  the end, though? The last review stops in mid-word.

Thank you, and I'm looking forward to meet you at the conference!
- Benja

Les Carr wrote:
> I am pleased to inform you that your submission
> (paper 192, Storm: Using P2P to make the desktop part of the Web) 
> has been accepted as a short paper for the ACM Hypertext 2003 conference.
> The quality has remained high with 25% of submissions being accepted,
> and despite the short timescale, most submissions were reviewed by
> four referees.
> 
> The timescale for the final version of your paper is very tight - we
> need to receive the final, camera-ready copy from you by Friday 20th June.
> When preparing the final version, you should take into account the referees'
> comments (below) and the formatting instructions on
> http://www.ht03.org/submissions.html
> 
> Thank you for your interest in HT03. We are anticipating a lively conference
> and look forward to your participation.
> ---
> Les Carr & Lynda Hardman
> HT03 PC CoChairs
> 
> ===================== REVIEWS ====================
> 
> Score: 8
> 	 An interesting look at a significant work in progress
> 
> Score: 7
> 	 Nice paper, look forward to seeing it in print. But before
> 	 it does, please don`t use the name "Storm" so many times
> 	 -sounds as if you are trying to sell us a product!  Also,
> 	 you have completely misunderstood what Schneier says about
> 	 cryptographic hashes in your reference to [10] - it is
> 	 *NOT* "practically impossible" that two documents will
> 	 _have_ the same hash (in fact, it is absolutely certainly
> 	 going to  happen), but rather, it is practically impossible
> 	 to _find_ any two documents that have the same hash. This
> 	 is the "collision resistance" property and it means that
> 	 is is hard to discover other documents with the same hash,
> 	 but it is quite easy for such to exist, purely on the
> 	 pigeon-hole principle (whether alternative documents with
> 	 the same hash are meaningful is another matter).
> 
> Score: 5
> 	 The motivation of `including the desktop` is valid, although
> 	 the UI changes that your suggested approach would require
> 	 are too numerous for such a change to be usable - a fact
> 	 to which you allude in Section 3.   It is also unclear
> 	 how using a straightforward hashing function, such as
> 	 SHA-1 or MD5, would suffice to cater for all resources in
> 	 the application space without collisions (I would argue
> 	 that it is insufficient).   Further, the choice of URN
> 	 naming is not new (e.g. eDonkey`s URN format for file
> 	 sharing), and there is no discussion of the impact of
> 	 moving from a structured naming scheme (of which users
> 	 may rightly or wrongly make assumptions regarding content)
> 	 to an opaque one.  Discussion of Storm`s security model
> 	 would help the reader assess the utility of having `all
> 	 documents stored on their local hard disk` included
> 	 automagically by the Storm system, especially in light of
> 	 recent discoveries of other p2p filesharing tools and
> 	 their somewhat questionable behaviour of file publication.
> 	 Th
> 
> 





From MAILER-DAEMON Sat Jun 14 16:31:15 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19RHgA-0007Co-S8
	for mharc-gzz-dev@gnu.org; Sat, 14 Jun 2003 16:31:14 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19RHg9-0007AN-NH
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 14 Jun 2003 16:31:13 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19RHg8-00079R-Fy
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 14 Jun 2003 16:31:12 -0400
Received: from verso.st.jyu.fi ([130.234.4.66])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19RHg8-000774-0t
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 14 Jun 2003 16:31:12 -0400
Received: from verso.st.jyu.fi (localhost.localdomain [127.0.0.1])
	by verso.st.jyu.fi (8.12.8/8.12.8) with ESMTP id h5EKV5ND020994;
	Sat, 14 Jun 2003 23:31:06 +0300
Received: from localhost (majukati@localhost)h5EKV5li020990;
	Sat, 14 Jun 2003 23:31:05 +0300
Date: Sat, 14 Jun 2003 23:31:05 +0300 (EEST)
From: Matti Katila <majukati@verso.st.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
In-Reply-To: <3EEB0185.6030600@gmx.de>
Message-ID: <Pine.LNX.4.44.0306142315040.31128-100000@verso.st.jyu.fi>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
cc: gzz-dev <gzz-dev@mail.freesoftware.fsf.org>
Subject: [Gzz] Re: [Gzz-commits] journals mudyc
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sat, 14 Jun 2003 20:31:14 -0000

On Sat, 14 Jun 2003, Benja Fallenstein wrote:
>> Module name:	journals
>> Changes by:	Matti Katila <majukati@cc.jyu.fi>	03/06/13 18:57:42
>> Modified files:
>> 	.              : mudyc 
>> Log message:
>> 	diibadaaba, i feel relaxed.. you should also take some vacation 
>>      without internet =)
> 
> Don't look at buoyoing yet, then. ;-) Or at least be prepared that I 
> broke text coloring (because RSTText doesn't do linebreaking and I 
> wanted linebreaking) and haven't found a good way to fix it, yet.

It shouldn't be too hard to fix, couple of lines only I think. I can start 
looking around it at night. There's a couple of callbacks only which need 
to be in order.
 
> I'm still mulling about a good way to select regions of text...

What part of it? Like the idea or with linebreaking? 
First we can do it with area selection code which is needed in PDFs to 
make xulinks (according to journals the area selecting of image span still 
remains in Tuomas's journal :-)


   -Matti




From MAILER-DAEMON Tue Jun 17 06:05:51 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19SDKs-0001of-9R
	for mharc-gzz-dev@gnu.org; Tue, 17 Jun 2003 06:05:06 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19SDKk-0001g8-Kx
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 06:04:58 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19SDK6-00017d-KD
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 06:04:21 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19SDIh-00083F-4Z; Tue, 17 Jun 2003 06:02:51 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5HA2ijX020884;
	Tue, 17 Jun 2003 13:02:45 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5HA2siL016712;
	Tue, 17 Jun 2003 13:02:54 +0300 (EETDST)
Date: Tue, 17 Jun 2003 13:02:54 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Message-ID: <20030617100254.GA6198@tarzan.it.jyu.fi>
References: <E19SAP3-00051U-Pn@subversions.gnu.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <E19SAP3-00051U-Pn@subversions.gnu.org>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5HA2ijX020884
cc: commitbot@gzz.info
cc: gzz-commits@nongnu.org
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Subject: [Gzz] Re: [Gzz-commits] storm/doc/pegboard/attacking_gisp--hemppah
	peg.rst
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 17 Jun 2003 10:05:04 -0000

On Tue, Jun 17, 2003 at 02:57:13AM -0400, Hermanni Hyyti=E4l=E4 wrote:
> CVSROOT:	/cvsroot/storm
> Module name:	storm
> Branch: =09
> Changes by:	Hermanni Hyyti=E4l=E4 <hemppah@cc.jyu.fi>	03/06/17 02:57:13
>=20
> Modified files:
> 	doc/pegboard/attacking_gisp--hemppah: peg.rst=20
>=20
> Log message:
> 	fix
>=20
> CVSWeb URLs:
> http://savannah.gnu.org/cgi-bin/viewcvs/storm/storm/doc/pegboard/attack=
ing_gisp--hemppah/peg.rst.diff?tr1=3D1.37&tr2=3D1.38&r1=3Dtext&r2=3Dtext
>=20
> Patches:
> Index: storm/doc/pegboard/attacking_gisp--hemppah/peg.rst
> diff -u storm/doc/pegboard/attacking_gisp--hemppah/peg.rst:1.37 storm/d=
oc/pegboard/attacking_gisp--hemppah/peg.rst:1.38
> --- storm/doc/pegboard/attacking_gisp--hemppah/peg.rst:1.37	Mon Jun 16 =
07:47:36 2003
> +++ storm/doc/pegboard/attacking_gisp--hemppah/peg.rst	Tue Jun 17 02:57=
:13 2003
> @@ -4,8 +4,8 @@
> =20
>  :Authors:  Hermanni Hyyti=E4l=E4
>  :Date-Created: 2003-06-05
> -:Last-Modified: $Date: 2003/06/16 11:47:36 $
> -:Revision: $Revision: 1.37 $
> +:Last-Modified: $Date: 2003/06/17 06:57:13 $
> +:Revision: $Revision: 1.38 $
>  :Status:   Incomplete
> =20
>  .. :Stakeholders:
> @@ -279,11 +279,11 @@
>    fail. =20
> =20
>    n peers, O(log n) path length, all peers perform n lookups. Then the=
 total
> -  number of packets is (n-1)((n-1)(log n))
> +  number of packets is n((n-1)(log n))
>   =20
>    n peers where fraction f are hostile, O(log n) path length, all peer=
s=20
>    perform n lookups. Then the total number of lost packets is=20
> -  [f^((log n)-1)]*[(n-1)((n-1)(log n))]
> +  [f^((log n)-1)]*[n((n-1)(log n))]
>   =20
Where are the sources / derivations for the formulas?

If you're using O(log n) for the path length, then you definitely
should not let "n-1" enter the formulas anywhere...

	Tuomas



From MAILER-DAEMON Tue Jun 17 07:27:22 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19SETH-0000Df-PV
	for mharc-gzz-dev@gnu.org; Tue, 17 Jun 2003 07:17:51 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19SEP7-0007iE-IJ
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 07:13:33 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19SEEr-0005kv-JC
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 07:02:58 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19SE6m-0005El-LA; Tue, 17 Jun 2003 06:54:36 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5HAsWjX024410;
	Tue, 17 Jun 2003 13:54:32 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5HAsgGK017341;
	Tue, 17 Jun 2003 13:54:42 +0300 (EETDST)
Date: Tue, 17 Jun 2003 13:54:41 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] Re: [Gzz-commits]
	storm/doc/pegboard/attacking_gisp--hemppah peg.rst
Message-ID: <20030617105441.GB6198@tarzan.it.jyu.fi>
References: <E19SAP3-00051U-Pn@subversions.gnu.org>
	<20030617100254.GA6198@tarzan.it.jyu.fi>
	<1055846612.26460.22.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1055846612.26460.22.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5HAsWjX024410
cc: commitbot@gzz.info
cc: gzz-commits@nongnu.org
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 17 Jun 2003 11:17:47 -0000

On Tue, Jun 17, 2003 at 01:43:32PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> > >    n peers, O(log n) path length, all peers perform n lookups. Then=
 the total
> > > -  number of packets is (n-1)((n-1)(log n))
> > > +  number of packets is n((n-1)(log n))
> > >   =20
> > >    n peers where fraction f are hostile, O(log n) path length, all =
peers=20
> > >    perform n lookups. Then the total number of lost packets is=20
> > > -  [f^((log n)-1)]*[(n-1)((n-1)(log n))]
> > > +  [f^((log n)-1)]*[n((n-1)(log n))]
> > >   =20
> > Where are the sources / derivations for the formulas?
>=20
> No sources, I derived those formulas myself, except that "f^((log n)-1)=
"
> (i.e., "probability of routing wrong") is derived from the "probability
> of routing successfully" formula: (1-f)^h-1.=20

> And the formula is an estimation (have to change the word in the PEG").

> > If you're using O(log n) for the path length, then you definitely
> > should not let "n-1" enter the formulas anywhere...
>=20
> Hm, "n-1" is in the formula since when a node queries the network, whic=
h
> has n nodes, no network messages are required to query "my local data".

Umm, I wasn't clear.

If you use path length =3D O(log n)

then you're already making a large approximation, and assuming n is fairl=
y
large.

In that case,=20

	n-1 \approx n=20

will hold and allows you to simplify your formulas nicely.

Ok?

> The idea behind the derivation is roughly:
>=20
> 1. "In a n node network, average lookup length is O(log n)".
> 2. "A node says: I will create (number of) n queries so that each query
> will reach other node in the network.=20

> In the end, for each node in the
> network, one query has reached the node".

This is not guaranteed for any n queries, right?

> 3. "Repeat step 2. for all n nodes".
>=20
> But I'm not sure if this is right thinking or not...

Sounds right but your presentation above is bizarre. Better use programma=
tic
notation:=20

For all nodes n:

	do foo...

or something like it. Don't repeat afterwards without saying before that
youre *going* to repeat.

	Tuomas



From MAILER-DAEMON Tue Jun 17 07:35:09 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19SEiI-00065P-54
	for mharc-gzz-dev@gnu.org; Tue, 17 Jun 2003 07:33:22 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19SEhy-0005sR-5X
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 07:33:02 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19SEY7-0001Gr-MC
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 07:22:52 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19SDw8-0003SA-O6
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 06:43:36 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5HAhWdc032431;	Tue, 17 Jun 2003 13:43:32 +0300
Subject: Re: [Gzz] Re: [Gzz-commits]
	storm/doc/pegboard/attacking_gisp--hemppah peg.rst
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030617100254.GA6198@tarzan.it.jyu.fi>
References: <E19SAP3-00051U-Pn@subversions.gnu.org>
	 <20030617100254.GA6198@tarzan.it.jyu.fi>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1055846612.26460.22.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 17 Jun 2003 13:43:32 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti2.jyu.fi id
	h5HAhWdc032431
cc: commitbot@gzz.info
cc: gzz-commits@nongnu.org
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 17 Jun 2003 11:33:16 -0000

On Tue, 2003-06-17 at 13:02, Tuomas Lukka wrote:
> On Tue, Jun 17, 2003 at 02:57:13AM -0400, Hermanni Hyyti=E4l=E4 wrote:
> > CVSROOT:	/cvsroot/storm
> > Module name:	storm
> > Branch: =09
> > Changes by:	Hermanni Hyyti=E4l=E4 <hemppah@cc.jyu.fi>	03/06/17 02:57:=
13
> >=20
> > Modified files:
> > 	doc/pegboard/attacking_gisp--hemppah: peg.rst=20
> >=20
> > Log message:
> > 	fix
> >=20
> > CVSWeb URLs:
> > http://savannah.gnu.org/cgi-bin/viewcvs/storm/storm/doc/pegboard/atta=
cking_gisp--hemppah/peg.rst.diff?tr1=3D1.37&tr2=3D1.38&r1=3Dtext&r2=3Dtex=
t
> >=20
> > Patches:
> > Index: storm/doc/pegboard/attacking_gisp--hemppah/peg.rst
> > diff -u storm/doc/pegboard/attacking_gisp--hemppah/peg.rst:1.37 storm=
/doc/pegboard/attacking_gisp--hemppah/peg.rst:1.38
> > --- storm/doc/pegboard/attacking_gisp--hemppah/peg.rst:1.37	Mon Jun 1=
6 07:47:36 2003
> > +++ storm/doc/pegboard/attacking_gisp--hemppah/peg.rst	Tue Jun 17 02:=
57:13 2003
> > @@ -4,8 +4,8 @@
> > =20
> >  :Authors:  Hermanni Hyyti=E4l=E4
> >  :Date-Created: 2003-06-05
> > -:Last-Modified: $Date: 2003/06/16 11:47:36 $
> > -:Revision: $Revision: 1.37 $
> > +:Last-Modified: $Date: 2003/06/17 06:57:13 $
> > +:Revision: $Revision: 1.38 $
> >  :Status:   Incomplete
> > =20
> >  .. :Stakeholders:
> > @@ -279,11 +279,11 @@
> >    fail. =20
> > =20
> >    n peers, O(log n) path length, all peers perform n lookups. Then t=
he total
> > -  number of packets is (n-1)((n-1)(log n))
> > +  number of packets is n((n-1)(log n))
> >   =20
> >    n peers where fraction f are hostile, O(log n) path length, all pe=
ers=20
> >    perform n lookups. Then the total number of lost packets is=20
> > -  [f^((log n)-1)]*[(n-1)((n-1)(log n))]
> > +  [f^((log n)-1)]*[n((n-1)(log n))]
> >   =20
> Where are the sources / derivations for the formulas?

No sources, I derived those formulas myself, except that "f^((log n)-1)"
(i.e., "probability of routing wrong") is derived from the "probability
of routing successfully" formula: (1-f)^h-1.=20

And the formula is an estimation (have to change the word in the PEG").

>=20
> If you're using O(log n) for the path length, then you definitely
> should not let "n-1" enter the formulas anywhere...

Hm, "n-1" is in the formula since when a node queries the network, which
has n nodes, no network messages are required to query "my local data".

The idea behind the derivation is roughly:

1. "In a n node network, average lookup length is O(log n)".
2. "A node says: I will create (number of) n queries so that each query
will reach other node in the network. In the end, for each node in the
network, one query has reached the node".
3. "Repeat step 2. for all n nodes".

But I'm not sure if this is right thinking or not...


-Hermanni








From MAILER-DAEMON Tue Jun 17 07:37:41 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19SEmT-00007y-1h
	for mharc-gzz-dev@gnu.org; Tue, 17 Jun 2003 07:37:41 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19SEm2-0008FK-Pf
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 07:37:14 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19SEm1-0008EF-4q
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 07:37:13 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19SEm0-0008DS-JX
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 07:37:12 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5HBbAqO005287;	Tue, 17 Jun 2003 14:37:10 +0300
Subject: Re: [Gzz] Re: [Gzz-commits]
	storm/doc/pegboard/attacking_gisp--hemppah peg.rst
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030617105441.GB6198@tarzan.it.jyu.fi>
References: <E19SAP3-00051U-Pn@subversions.gnu.org>
	 <20030617100254.GA6198@tarzan.it.jyu.fi>
	 <1055846612.26460.22.camel@io.it.jyu.fi>
	 <20030617105441.GB6198@tarzan.it.jyu.fi>
Content-Type: text/plain
Organization: 
Message-Id: <1055849830.26466.42.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 17 Jun 2003 14:37:10 +0300
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
cc: commitbot@gzz.info
cc: gzz-commits@nongnu.org
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 17 Jun 2003 11:37:39 -0000

On Tue, 2003-06-17 at 13:54, Tuomas Lukka wrote:
> 
> > > If you're using O(log n) for the path length, then you definitely
> > > should not let "n-1" enter the formulas anywhere...
> > 
> > Hm, "n-1" is in the formula since when a node queries the network, which
> > has n nodes, no network messages are required to query "my local data".
> 
> Umm, I wasn't clear.
> 
> If you use path length = O(log n)
> 
> then you're already making a large approximation, and assuming n is fairly
> large.
> 
> In that case, 
> 
> 	n-1 \approx n 
> 
> will hold and allows you to simplify your formulas nicely.
> 
> Ok?

I don't understand your "n-1 \approx n" syntax completely. Does that
mean that "n is an approximation, and...", or "Now, we replace this with
this, and...", ... ?

> 
> > The idea behind the derivation is roughly:
> > 
> > 1. "In a n node network, average lookup length is O(log n)".
> > 2. "A node says: I will create (number of) n queries so that each query
> > will reach other node in the network. 
> 
> > In the end, for each node in the
> > network, one query has reached the node".
> 
> This is not guaranteed for any n queries, right?

Yes, right. But in this case we can assume that this is an ideal network
that suits for our purposes and simplifies formula, right? In practice
of course, it's not guaranteed.



-Hermanni




From MAILER-DAEMON Tue Jun 17 08:33:28 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19SFdi-0004IC-K1
	for mharc-gzz-dev@gnu.org; Tue, 17 Jun 2003 08:32:42 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19SFdd-000470-OU
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 08:32:37 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19SFcg-0003W6-5q
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 08:31:38 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19SFcc-0003Ua-8n
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 17 Jun 2003 08:31:34 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5HCVWdc017386	for <gzz-dev@mail.freesoftware.fsf.org>;
	Tue, 17 Jun 2003 15:31:33 +0300
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Content-Type: text/plain
Organization: 
Message-Id: <1055853092.26460.67.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 17 Jun 2003 15:31:32 +0300
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Subject: [Gzz] Some P2P papers
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 17 Jun 2003 12:32:40 -0000

Hi,

(This email is primary targetted to those people in our research group
who are interested in P2P networks ;))

I found today good reading related to P2P networks:

"Routing Networks for Distributed Hash Tables"
http://www-cs.stanford.edu/~manku/papers/03podc-dht.pdf


"Novel Architectures for P2P Applications: the Continuous-Discrete
Approach"
http://www.wisdom.weizmann.ac.il/~uwieder/papers/spaa03camera.ps


"Deconstructing the Kazaa Network"
http://people.cs.uchicago.edu/~matei/PAPERS/kazaa.pdf



-Hermanni








From MAILER-DAEMON Wed Jun 18 09:50:05 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19SdJL-0005ef-Vo
	for mharc-gzz-dev@gnu.org; Wed, 18 Jun 2003 09:49:15 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19SdHw-0004tq-1E
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 18 Jun 2003 09:47:48 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19SdHX-0003v1-GF
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 18 Jun 2003 09:47:26 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19SdHP-0003LD-Ae; Wed, 18 Jun 2003 09:47:15 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5IDl5jX026138;
	Wed, 18 Jun 2003 16:47:05 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5IDlG5u003979;
	Wed, 18 Jun 2003 16:47:16 +0300 (EETDST)
Date: Wed, 18 Jun 2003 16:47:16 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] Re: [Gzz-commits]
	storm/doc/pegboard/attacking_gisp--hemppah peg.rst
Message-ID: <20030618134716.GB3767@tarzan.it.jyu.fi>
References: <E19SAP3-00051U-Pn@subversions.gnu.org>
	<20030617100254.GA6198@tarzan.it.jyu.fi>
	<1055846612.26460.22.camel@io.it.jyu.fi>
	<20030617105441.GB6198@tarzan.it.jyu.fi>
	<1055849830.26466.42.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1055849830.26466.42.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5IDl5jX026138
cc: commitbot@gzz.info
cc: gzz-commits@nongnu.org
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 18 Jun 2003 13:49:14 -0000

On Tue, Jun 17, 2003 at 02:37:10PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> On Tue, 2003-06-17 at 13:54, Tuomas Lukka wrote:
> >=20
> > > > If you're using O(log n) for the path length, then you definitely
> > > > should not let "n-1" enter the formulas anywhere...
> > >=20
> > > Hm, "n-1" is in the formula since when a node queries the network, =
which
> > > has n nodes, no network messages are required to query "my local da=
ta".
> >=20
> > Umm, I wasn't clear.
> >=20
> > If you use path length =3D O(log n)
> >=20
> > then you're already making a large approximation, and assuming n is f=
airly
> > large.
> >=20
> > In that case,=20
> >=20
> > 	n-1 \approx n=20
> >=20
> > will hold and allows you to simplify your formulas nicely.
> >=20
> > Ok?
>=20
> I don't understand your "n-1 \approx n" syntax completely. Does that
> mean that "n is an approximation, and...", or "Now, we replace this wit=
h
> this, and...", ... ?

\approx is the wiggly equals sign in LaTeX.

> > > The idea behind the derivation is roughly:
> > >=20
> > > 1. "In a n node network, average lookup length is O(log n)".
> > > 2. "A node says: I will create (number of) n queries so that each q=
uery
> > > will reach other node in the network.=20
> >=20
> > > In the end, for each node in the
> > > network, one query has reached the node".
> >=20
> > This is not guaranteed for any n queries, right?
>=20
> Yes, right. But in this case we can assume that this is an ideal networ=
k
> that suits for our purposes and simplifies formula, right? In practice
> of course, it's not guaranteed.

Even in an ideal network, if you send n queries, you can't expect=20
the targets to be on different hosts.

	Tuomas



From MAILER-DAEMON Sat Jun 21 03:55:53 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19TdE1-0004RM-AR
	for mharc-gzz-dev@gnu.org; Sat, 21 Jun 2003 03:55:53 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19TdDx-0004Nj-SX
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 21 Jun 2003 03:55:49 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19TdDw-0004L1-BE
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 21 Jun 2003 03:55:48 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19TdDv-0004Kn-Qo
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 21 Jun 2003 03:55:47 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5L7tg8m004877
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Sat, 21 Jun 2003 10:55:42 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5L7u16G025411
	for gzz-dev@mail.freesoftware.fsf.org;
	Sat, 21 Jun 2003 10:56:01 +0300 (EETDST)
Date: Sat, 21 Jun 2003 10:56:01 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Message-ID: <20030621075601.GC3767@tarzan.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Subject: [Gzz] 
 [oinas@stanford.edu: Fwd: JoDI CfP joint with NKOS Workshop ECDL2003 (fwd)]
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sat, 21 Jun 2003 07:55:50 -0000

Arrgh -- should we?

	Tuomas


----- Forwarded message from oinas@stanford.edu -----

X-Authentication-Warning: webmail.Stanford.EDU: wwwd set sender to oinas@stanford.edu using -f
To: tuomas.lukka@cc.jyu.fi
Subject: Fwd: JoDI CfP joint with NKOS Workshop ECDL2003 (fwd)
Date: Fri, 20 Jun 2003 11:19:53 -0700
From: oinas@stanford.edu
X-Mailer: Stanford Webmail v1.1.6 17-June-2002
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi

----- Forwarded message from Oinas-Kukkonen Harri <hok@tols16.oulu.fi> -
----
From: Oinas-Kukkonen Harri <hok@tols16.oulu.fi>
To: oinas@stanford.edu
Subject: JoDI CfP joint with NKOS Workshop ECDL2003 (fwd)
Date: Fri, 20 Jun 2003 21:14:20 +0300 (EEST)





---------- Forwarded message ----------
Date: Thu, 19 Jun 2003 13:28:11 +0100
From: JoDI Announcements <jodi@ecs.soton.ac.uk>
To: "hok@rieska.oulu.fi" <hok@tols17.oulu.fi>
Subject: JoDI CfP joint with NKOS Workshop ECDL2003

You have received this message as a registered user of the Journal of
Digital Information.

JoDI is pleased to announce a Call for Papers in cooperation with the 
NKOS
Workshop at ECDL2003. This is an extract from the full CfP:

"Presentations from the Workshop may be selected for consideration in a
forthcoming (early 2004) JoDI special issue on NKOS themes (edited by 
Koch,
Tudhope). Download statistics show that articles from the April 2001 
NKOS
JoDI issue (http://jodi.ecs.soton.ac.uk/?vol=1&iss=8), based on the ECDL
2000 NKOS Workshop, have been among the most read/downloaded articles in
JoDI (http://jodi.ecs.soton.ac.uk/sec.php3?content=editorial-v4i1)."

Deadline for papers: 1st July 2003.

The full Workshop announcement can be found at
http://www.glam.ac.uk/soc/research/hypermedia/NKOS-Workshop.php
--

The Journal of Digital Information is an electronic journal published 
only
via the Web. JoDI is currently free to users thanks to support from the
British Computer Society and Oxford University Press
http://jodi.ecs.soton.ac.uk/

If you do not wish to continue receiving these messages, you can
unsubscribe from JoDI by putting your email address into the form on 
this page
http://jodi.ecs.soton.ac.uk/register.php3
and pressing the button 'Remove me from list'.

----- End forwarded message -----

----- End forwarded message -----



From MAILER-DAEMON Mon Jun 23 08:44:58 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19UQfp-0007cj-S1
	for mharc-gzz-dev@gnu.org; Mon, 23 Jun 2003 08:43:53 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19UQfi-0007X6-TV
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 23 Jun 2003 08:43:46 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19UQcW-0006BC-Ah
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 23 Jun 2003 08:40:30 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19UQbH-0005fu-FW
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 23 Jun 2003 08:39:11 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5NCd68m007445
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Mon, 23 Jun 2003 15:39:06 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5NCdUx2018042
	for gzz-dev@mail.freesoftware.fsf.org;
	Mon, 23 Jun 2003 15:39:30 +0300 (EETDST)
Date: Mon, 23 Jun 2003 15:39:30 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Message-ID: <20030623123930.GA18027@tarzan.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Subject: [Gzz] PEG: alph imagespans get whole page
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 23 Jun 2003 12:43:51 -0000

=============================================================
PEG imagespan_getwholepage--tjl: Access entire page
=============================================================

:Author:   Tuomas J. Lukka
:Last-Modified: $Date: 2003/06/23 12:37:11 $
:Revision: $Revision: 1.1 $
:Status:   Current

Now that there are two distinct types of image spans: plain
``ImageSpan``s and ``PageImageSpan``s, it is not easy to get the entire
image that the current span is a subset of, as ``PageImageSpan``s require
different treatment. This is not good.

This PEG resolves the problem by adding a method to ``ImageSpan`` to 
get the entire 2D image the current span is a subset of.

Issues
======

- The name for the new method? We need to say it's the opposite of a subimage,
  or somehow express that it's a larger area.

    RESOLVED: getSuperImageSpan. It's not perfect but at least it expresses that.

Changes
=======

Into ``ImageSpan`` add the method ``getSuperImageSpan``::

    /** Get the whole contiguous 2D image that this span is a part of.
     */
    ImageSpan getSuperImageSpan();




From MAILER-DAEMON Tue Jun 24 05:08:46 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Ujmm-0002H8-Tt
	for mharc-gzz-dev@gnu.org; Tue, 24 Jun 2003 05:08:20 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Ujmk-0002B5-8z
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 24 Jun 2003 05:08:18 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Ujmi-00029l-Fz
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 24 Jun 2003 05:08:17 -0400
Received: from verso.st.jyu.fi ([130.234.4.66])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Ujmg-00021m-0e
	for gzz-dev@mail.freesoftware.fsf.org; Tue, 24 Jun 2003 05:08:14 -0400
Received: from verso.st.jyu.fi (localhost.localdomain [127.0.0.1])
	by verso.st.jyu.fi (8.12.8/8.12.8) with ESMTP id h5O989ND025615;
	Tue, 24 Jun 2003 12:08:09 +0300
Received: from localhost (majukati@localhost)h5O9896f025611;
	Tue, 24 Jun 2003 12:08:09 +0300
Date: Tue, 24 Jun 2003 12:08:08 +0300 (EEST)
From: Matti Katila <majukati@verso.st.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
In-Reply-To: <E19UVrX-0008Pc-FQ@subversions.gnu.org>
Message-ID: <Pine.LNX.4.44.0306241149450.31128-100000@verso.st.jyu.fi>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
cc: gzz-dev <gzz-dev@mail.freesoftware.fsf.org>
Subject: [Gzz] The common goal, Re: [Gzz-commits] fenfire/org/fenfire
 demo/buoyoing.py util/Contr...
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Tue, 24 Jun 2003 09:08:19 -0000


On Mon, 23 Jun 2003, Benja Fallenstein wrote:
> CVSROOT:	/cvsroot/fenfire
> Module name:	fenfire
> Branch: 	
> Changes by:	Benja Fallenstein <b.fallenstein@gmx.de>	03/06/23 14:16:19
> Log message:
> 	small steps in event handling

Hi Benja,

I think we should attempt to reach a common goal instead of just deleting 
things. I agree that my sketching was not perfect way to solve all 
problems with events but by not defining the way it wasn't fair by just 
remove the code. So can we solve this problem with a common goal and 
define the way we should do it? Then we success of taking the steps in 
event handling.


   -Matti




From MAILER-DAEMON Wed Jun 25 01:22:42 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19V2jx-0007fA-Si
	for mharc-gzz-dev@gnu.org; Wed, 25 Jun 2003 01:22:41 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19V2jv-0007et-VO
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 25 Jun 2003 01:22:39 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19V2ju-0007cb-Ey
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 25 Jun 2003 01:22:39 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19V2jt-0007ad-Tb
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 25 Jun 2003 01:22:38 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5P5MR8m028205;
	Wed, 25 Jun 2003 08:22:28 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5P5MTY3000379;
	Wed, 25 Jun 2003 08:22:29 +0300 (EETDST)
Date: Wed, 25 Jun 2003 08:22:29 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Matti Katila <majukati@verso.st.jyu.fi>
Subject: Re: [Gzz] The common goal, Re: [Gzz-commits] fenfire/org/fenfire
	demo/buoyoing.py util/Contr...
Message-ID: <20030625052229.GB296@tarzan.it.jyu.fi>
References: <E19UVrX-0008Pc-FQ@subversions.gnu.org>
	<Pine.LNX.4.44.0306241149450.31128-100000@verso.st.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Pine.LNX.4.44.0306241149450.31128-100000@verso.st.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
cc: gzz-dev <gzz-dev@mail.freesoftware.fsf.org>
cc: Benja Fallenstein <b.fallenstein@gmx.de>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 25 Jun 2003 05:22:40 -0000

On Tue, Jun 24, 2003 at 12:08:08PM +0300, Matti Katila wrote:
> 
> On Mon, 23 Jun 2003, Benja Fallenstein wrote:
> > CVSROOT:	/cvsroot/fenfire
> > Module name:	fenfire
> > Branch: 	
> > Changes by:	Benja Fallenstein <b.fallenstein@gmx.de>	03/06/23 14:16:19
> > Log message:
> > 	small steps in event handling
> 
> Hi Benja,
> 
> I think we should attempt to reach a common goal instead of just deleting 
> things. I agree that my sketching was not perfect way to solve all 
> problems with events but by not defining the way it wasn't fair by just 
> remove the code. So can we solve this problem with a common goal and 
> define the way we should do it? Then we success of taking the steps in 
> event handling.

Matti: How about a PEG then? Much better to do controversial design work
out in the open than by just coding...

That way you can get the "official stamp of approval" on your plan
and there will be no competing approach to worry about.

	Tuomas



From MAILER-DAEMON Wed Jun 25 20:19:18 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19VInK-00084m-JU
	for mharc-gzz-dev@gnu.org; Wed, 25 Jun 2003 18:31:14 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19VIge-0006dW-NM
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 25 Jun 2003 18:24:20 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19VIe0-00063E-Gx
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 25 Jun 2003 18:21:37 -0400
Received: from itu.st.jyu.fi ([130.234.4.65])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19VIdM-0005gA-Gl
	for gzz-dev@mail.freesoftware.fsf.org; Wed, 25 Jun 2003 18:20:56 -0400
Received: from itu.st.jyu.fi (localhost.localdomain [127.0.0.1])
	by itu.st.jyu.fi (8.12.8/8.12.8) with ESMTP id h5PMJkML018075
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Thu, 26 Jun 2003 01:19:46 +0300
Received: from localhost (hastrup@localhost)
	by itu.st.jyu.fi (8.12.8/8.12.8/Submit) with ESMTP id h5PMJkfG018071
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Thu, 26 Jun 2003 01:19:46 +0300
X-Authentication-Warning: itu.st.jyu.fi: hastrup owned process doing -bs
Date: Thu, 26 Jun 2003 01:19:46 +0300 (EEST)
From: Tuukka Hastrup <Tuukka.Hastrup@iki.fi>
X-X-Sender: hastrup@itu.st.jyu.fi
To: gzz-dev@mail.freesoftware.fsf.org
Message-ID: <Pine.LNX.4.44.0306260116470.18030-100000@itu.st.jyu.fi>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Gzz] Three weeks away
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Wed, 25 Jun 2003 22:31:12 -0000


I'll be disconnected for the following three weeks as I travel to Poland.


-- 
-- Trying to catch me? Just follow up my Electric Fingerprints
-- To help you: Tuukka.Hastrup@iki.fi
                http://www.iki.fi/Tuukka.Hastrup/
                IRCNet: Stugge/tuukkah @#pii,#fenfire,#ynna
                Jabber ID: tuukka@jabber.org, ICQ #11321669




From MAILER-DAEMON Fri Jun 27 07:44:00 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Vrdi-0001KY-61
	for mharc-gzz-dev@gnu.org; Fri, 27 Jun 2003 07:43:38 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Vrd9-00010W-Mz
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 27 Jun 2003 07:43:03 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Vqnh-0007II-3D
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 27 Jun 2003 06:49:54 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Vqbr-0005Xn-KA
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 27 Jun 2003 06:37:39 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5RAbYUA016886
	for <gzz-dev@mail.freesoftware.fsf.org>;
	Fri, 27 Jun 2003 13:37:34 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5RAbfxV028369
	for gzz-dev@mail.freesoftware.fsf.org;
	Fri, 27 Jun 2003 13:37:41 +0300 (EETDST)
Date: Fri, 27 Jun 2003 13:37:41 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Message-ID: <20030627103741.GA28346@tarzan.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Subject: [Gzz] PEG: ideas dirs (dartboard)
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 27 Jun 2003 11:43:35 -0000

=============================================================
PEG ideas_dirs--tjl: PEG-like directories for ideas
=============================================================

:Author:   Tuomas J. Lukka
:Last-Modified: $Date: 2003/06/27 10:35:55 $
:Revision: $Revision: 1.3 $
:Status:   Current

PEGs are a good format for discussing relatively finished ideas,
ready for implementation. Recent discussions with humppake and mudyc
have convinced me that we need also a freer forum which can be used
to package and label more fragmentary and unfinished ideas, to make
them easier to discuss.

Issues
======

Changes
=======

In addition to ``pegboard``, all subprojects will have 
a sibling ``dartboard`` subdirectory.  The ideas subdirectory will
contain subdirectories with names in a similar format to PEGs, e.g. ::
    
    type_and_connect--humppake
    elliptical_fillets--mudyc

The subdirectory should contain a main "idea.rst" file.

This is where the similarity with PEGs ends - there are no formatting
rules, no processes for accepting or rejecting ideas. The point is 
just to sketch out something and give it a label by which it can
be referred to.

If the author desires, he/she can follow the PEG conventions or ignore
them. 

Hand-drawn scanned sketches and images are encouraged.

If real review of an idea, for accepting or rejecting, is desired, a PEG
referring to the idea can be made.




From MAILER-DAEMON Fri Jun 27 08:59:54 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Vsll-0006Eu-52
	for mharc-gzz-dev@gnu.org; Fri, 27 Jun 2003 08:56:01 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19VshV-0004pm-AY
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 27 Jun 2003 08:51:37 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19VsaD-0002OY-4E
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 27 Jun 2003 08:44:06 -0400
Received: from posti2.jyu.fi ([130.234.4.33])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19VsXw-0001rS-Jl
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 27 Jun 2003 08:41:44 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5RCfgdc001770	for <gzz-dev@mail.freesoftware.fsf.org>;
	Fri, 27 Jun 2003 15:41:43 +0300
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
Content-Type: multipart/mixed; boundary="=-oxLkVUnxyl26EQhJwLsV"
Organization: 
Message-Id: <1056717702.26798.39.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 27 Jun 2003 15:41:42 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Subject: [Gzz] [Fwd: Bug report]
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 27 Jun 2003 12:55:59 -0000


--=-oxLkVUnxyl26EQhJwLsV
Content-Type: text/plain
Content-Transfer-Encoding: 7bit

FYI.


-Hermanni

--=-oxLkVUnxyl26EQhJwLsV
Content-Disposition: inline
Content-Description: Forwarded message - Bug report
Content-Type: message/rfc822

Subject: Bug report
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: kaffe@kaffe.org
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1056717658.26798.37.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 27 Jun 2003 15:40:59 +0300
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti2.jyu.fi id
	h5RCfgdc001770

Hi,

There seems to be a bug in Kaffe's java.io.StreamTokenizer package. The
bug occurs when I try to run Matt Welsh's free non-blocking IO package
(NBIO) (http://www.eecs.harvard.edu/~mdw/proj/java-nbio/).


More specifically, when I try test the installation by changing to the
sandStorm/test/basic directory and running (as the NBIO documentation
suggests):

  sandstorm sandstorm.cfg

Now, I get the following with Kaffe 1.1.0 (June 9, 2003):

- - -

Sandstorm v1.6 <mdw@cs.berkeley.edu>
  Starting at Fri Jun 27 12:12:57 GMT 2003

Token: # (type: -3)
Error opening configuration file 'sandstorm.cfg': java.io.IOException:
No section name found at line 1 of config file, read #
java.io.IOException: No section name found at line 1 of config file,
read #
        at java.lang.Throwable.fillInStackTrace(Throwable.java:native)
        at java.lang.Throwable.<init>(Throwable.java:44)
        at java.lang.Exception.<init>(Exception.java:24)
        at java.io.IOException.<init>(IOException.java:24)
        at
mdw.sandStorm.main.SandstormConfig$configSection.doRead(SandstormConfig.j=
ava:689)
        at
mdw.sandStorm.main.SandstormConfig$configSection.<init>(SandstormConfig.j=
ava:618)
        at
mdw.sandStorm.main.SandstormConfig.readFile(SandstormConfig.java:476)
        at
mdw.sandStorm.main.SandstormConfig.<init>(SandstormConfig.java:176)
        at mdw.sandStorm.main.Main.main(Main.java:88)
Usage:
        java mdw.sandStorm.main.Main [-profile] <configfile> [initargs]

- - -

With IBM's JDK version 1.3 the following appears:

- - -


Sandstorm v1.6 <mdw@cs.berkeley.edu>
  Starting at Fri Jun 27 15:14:55 GMT+03:00 2003

Token: <sandstorm> (type: -3)
Token: <global> (type: -3)
Token: <threadPool> (type: -3)
Token: <sizeController> (type: -3)
Token: <aSocket> (type: -3)
Token: <rateController> (type: -3)
Token: <aDisk> (type: -3)
Token: <stages> (type: -3)
Token: <TimerStage> (type: -3)
Token: <initargs> (type: -3)
Token: <GenStage1> (type: -3)
Token: <initargs> (type: -3)
Token: <SinkStage> (type: -3)
ThreadPoolController: Started, delay 2000 ms, threshold 1000,
autoMaxDetect false
Sandstorm: Starting aSocket layer
SelectSet: Using poll(2)
SelectSource created, do_balance =3D true
TP <aSocket ReadStage>: Adding 1 threads to pool, size 1
TP <aSocket ReadStage>: Starting 1 threads, maxBatch=3D-1

...

- - -

And finally with Sun's JDK 1.4 (1.4.1_01):

- - -
Sandstorm v1.6 <mdw@cs.berkeley.edu>
  Starting at Fri Jun 27 15:17:30 EEST 2003

Token: <sandstorm> (type: -3)
Token: <global> (type: -3)
Token: <threadPool> (type: -3)
Token: <sizeController> (type: -3)
Token: <aSocket> (type: -3)
Token: <rateController> (type: -3)
Token: <aDisk> (type: -3)
Token: <stages> (type: -3)
Token: <TimerStage> (type: -3)
Token: <initargs> (type: -3)
Token: <GenStage1> (type: -3)
Token: <initargs> (type: -3)
Token: <SinkStage> (type: -3)
ThreadPoolController: Started, delay 2000 ms, threshold 1000,
autoMaxDetect false
Sandstorm: Starting aSocket layer
SelectSet: Using poll(2)
SelectSource created, do_balance =3D true
TP <aSocket ReadStage>: Adding 1 threads to pool, size 1
TP <aSocket ReadStage>: Starting 1 threads, maxBatch=3D-1

...

- - -

The start of the sandstorm.cfg looks like this:

- - -

# Example Sandstorm configuration file.
#
# The '#' starts a comment which extends to the end of the line
# This file uses an XML-like format consisting of nested sections.
#=20
# Most of the sections of this file are optional. The complete set of
# options is given here just to document their use, and leaving options=20
# unspecified causes them to use their default values. In general it is
# a good idea to just use the defaults.
#
# Special directives are indicated using <! ... > tags. Presently only
# one special directive is supported: <!include filename> which causes
# the named file to be included. Nested <!include> directives are
# supported.=20

# The outermost section in the file must be called 'sandstorm'.
<sandstorm>

  # Global options
  <global>


- - -

The tokenizer is initialized in
sandStorm.main.SandstormConfig$configSection (starts from line 610) like
this:

- - -
...

tok =3D new StreamTokenizer(in);
tok.resetSyntax();
tok.wordChars((char)0, (char)255);
tok.whitespaceChars('\u0000', '\u0020');
tok.commentChar('#');
tok.eolIsSignificant(true);

...

- - -

As seen, the '#' character is defined as the comment token. However,
based on Sandstorm's outputs above, it looks like that Kaffe's
StreamTokenizer does not handle the '#' charackter properly, i.e.,
StreamTokenizer threats the '#' character at line 1 as a "WORD" token
(TT_WORD =3D -3). Also, when I remove the comment lines from the
sandstorm.cfg file, the test will continue further.

Please notice that the original source code of NBIO distribution does
not include the "Token: <token> (type: <integer>)" debugging
information; I modified the original source code of NBIO in order make
this bug report more straightforward.

The NBIO/Sandstorm relase is 1.5 which is distributed with the Tapestry
1.0 release (http://www.cs.berkeley.edu/~ravenben/tapestry/).



Regards,
Hermanni Hyyti=E4l=E4

--=-oxLkVUnxyl26EQhJwLsV--




From MAILER-DAEMON Fri Jun 27 10:28:16 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Vtwr-0003U4-PQ
	for mharc-gzz-dev@gnu.org; Fri, 27 Jun 2003 10:11:33 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Vtw9-00038l-3I
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 27 Jun 2003 10:10:49 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Vtqm-0001uC-5t
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 27 Jun 2003 10:05:17 -0400
Received: from imap.gmx.net ([213.165.64.20] helo=mail.gmx.net)
	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19VtkX-0007jx-Nk
	for gzz-dev@mail.freesoftware.fsf.org; Fri, 27 Jun 2003 09:58:49 -0400
Received: (qmail 16236 invoked by uid 65534); 27 Jun 2003 13:58:47 -0000
Received: from pD95882C3.dip.t-dialin.net (EHLO gmx.de) (217.88.130.195)
  by mail.gmx.net (mp023) with SMTP; 27 Jun 2003 15:58:47 +0200
Message-ID: <3EFC4D5A.9050304@gmx.de>
Date: Fri, 27 Jun 2003 15:57:46 +0200
From: Benja Fallenstein <b.fallenstein@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.3.1) Gecko/20030618 Debian/1.3.1-3
X-Accept-Language: en
MIME-Version: 1.0
To: Tuomas Lukka <lukka@mit.jyu.fi>
Subject: Re: [Gzz] PEG: ideas dirs (dartboard)
References: <20030627103741.GA28346@tarzan.it.jyu.fi>
In-Reply-To: <20030627103741.GA28346@tarzan.it.jyu.fi>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Fri, 27 Jun 2003 14:11:32 -0000


Posting to the mailing list as soon as the idea is in a state where 
others can understand it should also be encouraged (but not mandated).

- Benja

Tuomas Lukka wrote:
> =============================================================
> PEG ideas_dirs--tjl: PEG-like directories for ideas
> =============================================================
> 
> :Author:   Tuomas J. Lukka
> :Last-Modified: $Date: 2003/06/27 10:35:55 $
> :Revision: $Revision: 1.3 $
> :Status:   Current
> 
> PEGs are a good format for discussing relatively finished ideas,
> ready for implementation. Recent discussions with humppake and mudyc
> have convinced me that we need also a freer forum which can be used
> to package and label more fragmentary and unfinished ideas, to make
> them easier to discuss.
> 
> Issues
> ======
> 
> Changes
> =======
> 
> In addition to ``pegboard``, all subprojects will have 
> a sibling ``dartboard`` subdirectory.  The ideas subdirectory will
> contain subdirectories with names in a similar format to PEGs, e.g. ::
>     
>     type_and_connect--humppake
>     elliptical_fillets--mudyc
> 
> The subdirectory should contain a main "idea.rst" file.
> 
> This is where the similarity with PEGs ends - there are no formatting
> rules, no processes for accepting or rejecting ideas. The point is 
> just to sketch out something and give it a label by which it can
> be referred to.
> 
> If the author desires, he/she can follow the PEG conventions or ignore
> them. 
> 
> Hand-drawn scanned sketches and images are encouraged.
> 
> If real review of an idea, for accepting or rejecting, is desired, a PEG
> referring to the idea can be made.
> 
> 
> 
> _______________________________________________
> Gzz-dev mailing list
> Gzz-dev@nongnu.org
> http://mail.nongnu.org/mailman/listinfo/gzz-dev
> 
> 





From MAILER-DAEMON Sat Jun 28 08:56:59 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19WFGE-0000wB-Lt
	for mharc-gzz-dev@gnu.org; Sat, 28 Jun 2003 08:56:58 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19WFGC-0000pi-4h
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 28 Jun 2003 08:56:56 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19WFGA-0000lZ-MV
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 28 Jun 2003 08:56:55 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19WFGA-0000Y0-5x
	for gzz-dev@mail.freesoftware.fsf.org; Sat, 28 Jun 2003 08:56:54 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5SCumUA026832;
	Sat, 28 Jun 2003 15:56:48 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5SCuwlD005598;
	Sat, 28 Jun 2003 15:56:58 +0300 (EETDST)
Date: Sat, 28 Jun 2003 15:56:58 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] [Fwd: Bug report]
Message-ID: <20030628125658.GA5573@tarzan.it.jyu.fi>
References: <1056717702.26798.39.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1056717702.26798.39.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5SCumUA026832
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sat, 28 Jun 2003 12:56:56 -0000

On Fri, Jun 27, 2003 at 03:41:42PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> FYI.
>=20
>=20
> -Hermanni

Content-Description: Forwarded message - Bug report
> Subject: Bug report
> From: Hermanni Hyyti=E4l=E4 <hemppah@cc.jyu.fi>
> To: kaffe@kaffe.org
> Organization:=20
> X-Mailer: Ximian Evolution 1.2.3=20
> Date: 27 Jun 2003 15:40:59 +0300
> X-MIME-Autoconverted: from 8bit to quoted-printable by posti2.jyu.fi id
> 	h5RCfgdc001770
>=20
> Hi,
>=20
> There seems to be a bug in Kaffe's java.io.StreamTokenizer package. The
> bug occurs when I try to run Matt Welsh's free non-blocking IO package
> (NBIO) (http://www.eecs.harvard.edu/~mdw/proj/java-nbio/).
>=20
>=20
> More specifically, when I try test the installation by changing to the
> sandStorm/test/basic directory and running (as the NBIO documentation
> suggests):
>=20

Umm - a comment about your bugreport: you should really narrow it down
to a really short java program that you can send them, i.e. 10 lines of=20
java, showing that on SUN's jdk it produces output XYZ and on kaffe, outp=
ut FOO
and that FOO is wrong according to the docs.

Did you also mention the kaffe version you discovered this on?

	Tuomas




From MAILER-DAEMON Sun Jun 29 01:07:30 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19WUPI-0007IN-3K
	for mharc-gzz-dev@gnu.org; Sun, 29 Jun 2003 01:07:20 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19WUOp-0006wP-0b
	for gzz-dev@mail.freesoftware.fsf.org; Sun, 29 Jun 2003 01:06:51 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19WUOe-0006PU-JV
	for gzz-dev@mail.freesoftware.fsf.org; Sun, 29 Jun 2003 01:06:42 -0400
Received: from c-24-98-47-52.atl.client2.attbi.com ([24.98.47.52]
	helo=61.18.88.249)	by monty-python.gnu.org with smtp (Exim 4.20)
	id 19WUOb-0005cu-Eu	for gzz-dev@mail.freesoftware.fsf.org;
	Sun, 29 Jun 2003 01:06:37 -0400
Received: from unknown (185.176.53.24) by rly-yk05.mx.aol.com with local;
	Jun, 29 2003 04:44:03 -0300
Received: from [41.237.71.37] by f64.law4.hotmail.com with NNFMP;
	Jun, 29 2003 04:02:19 +0600
From: Amazing Pranks <sxpsjensen@aol.com>
To: gzz-dev@mail.freesoftware.fsf.org
Sender: Amazing Pranks <sxpsjensen@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Date: Sun, 29 Jun 2003 06:04:32 +0100
X-Mailer: Microsoft Outlook Build 10.0.2616
Message-Id: <E19WUOb-0005cu-Eu@monty-python.gnu.org>
Subject: [Gzz] The funny prank people
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sun, 29 Jun 2003 05:07:17 -0000

To Inflict Crazy Prank Calls please visit this site:

http://www.geocities.com/rdjh546u465jj/





From MAILER-DAEMON Sun Jun 29 03:39:12 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19WWjY-0005lQ-2w
	for mharc-gzz-dev@gnu.org; Sun, 29 Jun 2003 03:36:24 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19WWir-0005cR-8L
	for gzz-dev@mail.freesoftware.fsf.org; Sun, 29 Jun 2003 03:35:41 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19WWh3-0005MR-LA
	for gzz-dev@mail.freesoftware.fsf.org; Sun, 29 Jun 2003 03:33:51 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19WWfY-0004zS-UM
	for gzz-dev@mail.freesoftware.fsf.org; Sun, 29 Jun 2003 03:32:17 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5T7WDUA002690;
	Sun, 29 Jun 2003 10:32:13 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5T7WPf6009458;
	Sun, 29 Jun 2003 10:32:25 +0300 (EETDST)
Date: Sun, 29 Jun 2003 10:32:25 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Benja Fallenstein <b.fallenstein@gmx.de>
Subject: Re: [Gzz] PEG: ideas dirs (dartboard)
Message-ID: <20030629073225.GC5573@tarzan.it.jyu.fi>
References: <20030627103741.GA28346@tarzan.it.jyu.fi>
	<3EFC4D5A.9050304@gmx.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3EFC4D5A.9050304@gmx.de>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Sun, 29 Jun 2003 07:36:21 -0000

On Fri, Jun 27, 2003 at 03:57:46PM +0200, Benja Fallenstein wrote:
> 
> Posting to the mailing list as soon as the idea is in a state where 
> others can understand it should also be encouraged (but not mandated).
> 
> - Benja
> 
> Tuomas Lukka wrote:
> >=============================================================
> >PEG ideas_dirs--tjl: PEG-like directories for ideas
> >=============================================================
> >


Added.

	Tuomas



From MAILER-DAEMON Mon Jun 30 06:00:24 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19WvRw-0005tI-0a
	for mharc-gzz-dev@gnu.org; Mon, 30 Jun 2003 05:59:52 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19WvQv-0005Sa-86
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 05:58:49 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19WvNH-0003wp-0b
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 05:55:04 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19WvMV-0003jy-2V
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 05:54:15 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5U9sCqO031744;	Mon, 30 Jun 2003 12:54:12 +0300
Subject: Re: [Gzz] [Fwd: Bug report]
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030628125658.GA5573@tarzan.it.jyu.fi>
References: <1056717702.26798.39.camel@io.it.jyu.fi>
	 <20030628125658.GA5573@tarzan.it.jyu.fi>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1056966852.15436.36.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 30 Jun 2003 12:54:12 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti1.jyu.fi id
	h5U9sCqO031744
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 30 Jun 2003 09:59:50 -0000

On Sat, 2003-06-28 at 15:56, Tuomas Lukka wrote:
> On Fri, Jun 27, 2003 at 03:41:42PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> > FYI.
> >=20
> >=20
> > -Hermanni
>=20
> Content-Description: Forwarded message - Bug report
> > Subject: Bug report
> > From: Hermanni Hyyti=E4l=E4 <hemppah@cc.jyu.fi>
> > To: kaffe@kaffe.org
> > Organization:=20
> > X-Mailer: Ximian Evolution 1.2.3=20
> > Date: 27 Jun 2003 15:40:59 +0300
> > X-MIME-Autoconverted: from 8bit to quoted-printable by posti2.jyu.fi =
id
> > 	h5RCfgdc001770
> >=20
> > Hi,
> >=20
> > There seems to be a bug in Kaffe's java.io.StreamTokenizer package. T=
he
> > bug occurs when I try to run Matt Welsh's free non-blocking IO packag=
e
> > (NBIO) (http://www.eecs.harvard.edu/~mdw/proj/java-nbio/).
> >=20
> >=20
> > More specifically, when I try test the installation by changing to th=
e
> > sandStorm/test/basic directory and running (as the NBIO documentation
> > suggests):
> >=20
>=20
> Umm - a comment about your bugreport: you should really narrow it down
> to a really short java program that you can send them, i.e. 10 lines of=
=20
> java, showing that on SUN's jdk it produces output XYZ and on kaffe, ou=
tput FOO
> and that FOO is wrong according to the docs.

IMO, partly true. I really wanted describe how&when&why the bug was
discovered. I think adding 10 lines of simple java code is *optional*
(not mandatory) in the bug report since the authors of code can *verify*
the bug report's result by coding those 10 lines easily.

I think that how&when&why concept is more important in a bug report than
10 lines of code representing a wrong functionality. But, OTOH, 10 lines
can be included if one wants so ;).


> Did you also mention the kaffe version you discovered this on?

Yes.


-Hermanni




From MAILER-DAEMON Mon Jun 30 06:16:31 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19WvgX-0002dM-MP
	for mharc-gzz-dev@gnu.org; Mon, 30 Jun 2003 06:14:57 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19WveE-0001dy-8I
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 06:12:34 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19WvYU-0008G2-5z
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 06:06:40 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19WvYC-00086w-Hh
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 06:06:20 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5UA6EUA003462;
	Mon, 30 Jun 2003 13:06:14 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5UA6TbX018072;
	Mon, 30 Jun 2003 13:06:29 +0300 (EETDST)
Date: Mon, 30 Jun 2003 13:06:28 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] [Fwd: Bug report]
Message-ID: <20030630100628.GD5573@tarzan.it.jyu.fi>
References: <1056717702.26798.39.camel@io.it.jyu.fi>
	<20030628125658.GA5573@tarzan.it.jyu.fi>
	<1056966852.15436.36.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1056966852.15436.36.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5UA6EUA003462
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 30 Jun 2003 10:14:56 -0000

On Mon, Jun 30, 2003 at 12:54:12PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> > Umm - a comment about your bugreport: you should really narrow it dow=
n
> > to a really short java program that you can send them, i.e. 10 lines =
of=20
> > java, showing that on SUN's jdk it produces output XYZ and on kaffe, =
output FOO
> > and that FOO is wrong according to the docs.
>=20
> IMO, partly true. I really wanted describe how&when&why the bug was
> discovered. I think adding 10 lines of simple java code is *optional*
> (not mandatory) in the bug report since the authors of code can *verify=
*
> the bug report's result by coding those 10 lines easily.
>=20
> I think that how&when&why concept is more important in a bug report tha=
n
> 10 lines of code representing a wrong functionality. But, OTOH, 10 line=
s
> can be included if one wants so ;).

I disagree completely. This is an arrogant attitude which I really wish
people who send me bug reports didn't take.

Here's again the numbers stuff: if the developers get 100 bug reports
a day, do you think they prefer to read through the short story of discov=
ery
or just fix it?

Most bug reports are false. If the authors had to code 10 lines for each =
bug report
to make sure that it's true, or install large packages to test it &c, the=
n
it would just not be feasible - they'd spend all their time doing that.

Doing the 10 lines is very little work for you, because you now know what
it's about but for them it's much more since they have to start thinking =
about this --
while not even being sure whether you found a bug or have just misunderst=
ood this.

Basically, if I get a bug report, I'd like to see

1) what the bug is
2) why it is a bug
3) example code so that I can *duplicate* the bug easily on my machine/en=
vironment
   or see that my later changes have already fixed it
4) exact configuration
5) (optional but strongly encouraged) What this bug affects - your intern=
al code /=20
   some important package &c
6) (optional) potential fix

Adding the java code is optional only if you cannot, for some reason,=20
make that code (i.e. you haven't found a short program that duplicates
the problem or you are not a coder).

	Tuomas



From MAILER-DAEMON Mon Jun 30 07:33:15 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Wwb7-0005XQ-Vr
	for mharc-gzz-dev@gnu.org; Mon, 30 Jun 2003 07:13:25 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19WwQH-0004SN-Tj
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 07:02:13 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19WwPg-0004F8-6Y
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 07:01:38 -0400
Received: from posti1.jyu.fi ([130.234.4.32])
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19Ww5Y-0000jA-K4
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 06:40:48 -0400
Received: from io.it.jyu.fi (hemppah@io.it.jyu.fi [130.234.160.112])
	h5UAelqO010138;	Mon, 30 Jun 2003 13:40:47 +0300
Subject: Re: [Gzz] [Fwd: Bug report]
From: Hermanni =?ISO-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
To: Tuomas Lukka <lukka@mit.jyu.fi>
In-Reply-To: <20030630100628.GD5573@tarzan.it.jyu.fi>
References: <1056717702.26798.39.camel@io.it.jyu.fi>
	 <20030628125658.GA5573@tarzan.it.jyu.fi>
	 <1056966852.15436.36.camel@io.it.jyu.fi>
	 <20030630100628.GD5573@tarzan.it.jyu.fi>
Content-Type: text/plain; charset=ISO-8859-1
Organization: 
Message-Id: <1056969646.15436.69.camel@io.it.jyu.fi>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.3 
Date: 30 Jun 2003 13:40:46 +0300
X-Virus-Scanned: by amavisd-milter (http://amavis.org/)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by posti1.jyu.fi id
	h5UAelqO010138
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 30 Jun 2003 11:13:24 -0000

On Mon, 2003-06-30 at 13:06, Tuomas Lukka wrote:
> On Mon, Jun 30, 2003 at 12:54:12PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> > > Umm - a comment about your bugreport: you should really narrow it d=
own
> > > to a really short java program that you can send them, i.e. 10 line=
s of=20
> > > java, showing that on SUN's jdk it produces output XYZ and on kaffe=
, output FOO
> > > and that FOO is wrong according to the docs.
> >=20
> > IMO, partly true. I really wanted describe how&when&why the bug was
> > discovered. I think adding 10 lines of simple java code is *optional*
> > (not mandatory) in the bug report since the authors of code can *veri=
fy*
> > the bug report's result by coding those 10 lines easily.
> >=20
> > I think that how&when&why concept is more important in a bug report t=
han
> > 10 lines of code representing a wrong functionality. But, OTOH, 10 li=
nes
> > can be included if one wants so ;).
>=20
> I disagree completely. This is an arrogant attitude which I really wish
> people who send me bug reports didn't take.
>=20
> Here's again the numbers stuff: if the developers get 100 bug reports
> a day, do you think they prefer to read through the short story of disc=
overy
> or just fix it?
>=20
> Most bug reports are false. If the authors had to code 10 lines for eac=
h bug report
> to make sure that it's true, or install large packages to test it &c, t=
hen
> it would just not be feasible - they'd spend all their time doing that.

I don't see anything wrong in this. I think that one important role of a
developer is also to *serve* end users among *developing* the software;
doing software is not only *developing* software, it's also fulfilling
end users' different needs (e.g. by fixing bugs/verifying if something
is a bug). And I think this the *responsibility* what a developer (the
*author* of software) takes when she/he releases a software, whether its
open source or non-open source software.

And what is important that and end user may be also a developer.

>=20
> Doing the 10 lines is very little work for you, because you now know wh=
at
> it's about but for them it's much more since they have to start thinkin=
g about this --
> while not even being sure whether you found a bug or have just misunder=
stood this.

> Basically, if I get a bug report, I'd like to see
>=20
> 1) what the bug is
> 2) why it is a bug
> 3) example code so that I can *duplicate* the bug easily on my machine/=
environment
>    or see that my later changes have already fixed it
> 4) exact configuration
> 5) (optional but strongly encouraged) What this bug affects - your inte=
rnal code /=20
>    some important package &c
> 6) (optional) potential fix
>=20
> Adding the java code is optional only if you cannot, for some reason,=20
> make that code (i.e. you haven't found a short program that duplicates
> the problem or you are not a coder).

Ok, in my case, the 10 lines of code would be feasible since I *can*
code (i.e. I'm (also) a developer). So I agree this in my case.=20

But in case of a regular end user/beginner programmer (as you mention
it), it is almost always impossible to create a simple source code which
would reveal the bug. Also, many end users (even in open source
community) are not able to do a "complete" bug reports since they don't
necessiraly know even their *exact* configuration, to say nothing of 5)
and 6).

Finally, there seems to be clear difference between non-open source
community and open source community with bug reporting in general ;).
While in non-open source community people sends only "this does not
work" messages with *automatically generated* configuration reports, in
open source community it is expected that (almost) all end users are
also developers &c: when making a bug report it is assumed that you
would provide "complete" bug report (even) containing a potential fix.

But hey, isn't this the thing why open source software do exist ? ;)


-Hermanni






From MAILER-DAEMON Mon Jun 30 08:42:38 2003
Received: from list by monty-python.gnu.org with archive (Exim 4.20)
	id 19Wxyo-0006mC-HL
	for mharc-gzz-dev@gnu.org; Mon, 30 Jun 2003 08:41:58 -0400
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.20)
	id 19Wxlu-0003FL-9P
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 08:28:38 -0400
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.20)
	id 19Wxgh-0002Iy-7e
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 08:23:21 -0400
Received: from posti.it.jyu.fi ([130.234.160.33] helo=mailhost.it.jyu.fi)
	by monty-python.gnu.org with esmtp (Exim 4.20)
	id 19WxZ8-00009K-Gc
	for gzz-dev@mail.freesoftware.fsf.org; Mon, 30 Jun 2003 08:15:26 -0400
Received: from tarzan.mit.jyu.fi (tarzan.private [172.16.160.40])
	by mailhost.it.jyu.fi (8.12.9/8.12.9) with ESMTP id h5UCFHUA013057;
	Mon, 30 Jun 2003 15:15:17 +0300 (EETDST)
Received: (from lukka@localhost)
	by tarzan.mit.jyu.fi (8.12.9/8.12.9/Submit) id h5UCFWBY019310;
	Mon, 30 Jun 2003 15:15:32 +0300 (EETDST)
Date: Mon, 30 Jun 2003 15:15:32 +0300
From: Tuomas Lukka <lukka@mit.jyu.fi>
To: Hermanni =?iso-8859-1?Q?Hyyti=E4l=E4?= <hemppah@cc.jyu.fi>
Subject: Re: [Gzz] [Fwd: Bug report]
Message-ID: <20030630121532.GE5573@tarzan.it.jyu.fi>
References: <1056717702.26798.39.camel@io.it.jyu.fi>
	<20030628125658.GA5573@tarzan.it.jyu.fi>
	<1056966852.15436.36.camel@io.it.jyu.fi>
	<20030630100628.GD5573@tarzan.it.jyu.fi>
	<1056969646.15436.69.camel@io.it.jyu.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <1056969646.15436.69.camel@io.it.jyu.fi>
User-Agent: Mutt/1.5.4i
X-Virus-Scanned-By: amavisd-milter at it.jyu.fi
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mailhost.it.jyu.fi id
	h5UCFHUA013057
cc: gzz developers list <gzz-dev@mail.freesoftware.fsf.org>
X-BeenThere: gzz-dev@nongnu.org
X-Mailman-Version: 2.1b5
Precedence: list
List-Id: Gzz developers <gzz-dev.nongnu.org>
List-Help: <mailto:gzz-dev-request@nongnu.org?subject=help>
List-Post: <mailto:gzz-dev@nongnu.org>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=subscribe>
List-Archive: <http://mail.gnu.org/pipermail/gzz-dev>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/gzz-dev>,
	<mailto:gzz-dev-request@nongnu.org?subject=unsubscribe>
X-List-Received-Date: Mon, 30 Jun 2003 12:41:56 -0000

On Mon, Jun 30, 2003 at 01:40:46PM +0300, Hermanni Hyyti=E4l=E4 wrote:
> On Mon, 2003-06-30 at 13:06, Tuomas Lukka wrote:
> > On Mon, Jun 30, 2003 at 12:54:12PM +0300, Hermanni Hyyti=E4l=E4 wrote=
:
> > > > Umm - a comment about your bugreport: you should really narrow it=
 down
> > > > to a really short java program that you can send them, i.e. 10 li=
nes of=20
> > > > java, showing that on SUN's jdk it produces output XYZ and on kaf=
fe, output FOO
> > > > and that FOO is wrong according to the docs.
> > >=20
> > > IMO, partly true. I really wanted describe how&when&why the bug was
> > > discovered. I think adding 10 lines of simple java code is *optiona=
l*
> > > (not mandatory) in the bug report since the authors of code can *ve=
rify*
> > > the bug report's result by coding those 10 lines easily.
> > >=20
> > > I think that how&when&why concept is more important in a bug report=
 than
> > > 10 lines of code representing a wrong functionality. But, OTOH, 10 =
lines
> > > can be included if one wants so ;).
> >=20
> > I disagree completely. This is an arrogant attitude which I really wi=
sh
> > people who send me bug reports didn't take.
> >=20
> > Here's again the numbers stuff: if the developers get 100 bug reports
> > a day, do you think they prefer to read through the short story of di=
scovery
> > or just fix it?
> >=20
> > Most bug reports are false. If the authors had to code 10 lines for e=
ach bug report
> > to make sure that it's true, or install large packages to test it &c,=
 then
> > it would just not be feasible - they'd spend all their time doing tha=
t.
>=20
> I don't see anything wrong in this.=20

Really?  I'm astonished.

> I think that one important role of a
> developer is also to *serve* end users among *developing* the software;
> doing software is not only *developing* software, it's also fulfilling
> end users' different needs (e.g. by fixing bugs/verifying if something
> is a bug).=20

This would be the right attitude if the developers were really working
for you and you were paying them money.

However, this is not the case.

Here, the deal is

	- they put out free software
	- they are often also willing to fix bugs

If your side is

	"I don't want to bother to write a test case - I'll just
	send them some notes about how I found a bug and some notes
	about what the bug itself is"

then the deal is a little off balance, right?

The correct and, most of all, **polite** way is

	"Hmm, I've found a bug - what can I do to make it easiest
	for them to fix it within reason?"

Within reason: If you're not familiar with the code, you don't need to wr=
ite
an actual patch, but if you're a coder, a reproducible test case if
of very little bother to you and of enormous value to the person
receiving the report.

There are free software projects that have ENDED because people sent in
vague and bad bug reports and the developers got tired of doing the proje=
ct --
very understandably.

> And I think this the *responsibility* what a developer (the
> *author* of software) takes when she/he releases a software, whether it=
s
> open source or non-open source software.

No. You can't go on assigning other people responsibilities.

If they are willing, out of their own goodwill, continue development
and fix bugs, you should be grateful to them and make their life easier.

You're forgetting the cardinal rule of RFC:s:

	be strict in what you produce, and liberal in what you accept

It's right to *feel* the responsibility and act accordingly. It's NOT RIG=
HT
to *EXPECT* it from others about whose situation you have no knowledge.

> And what is important that and end user may be also a developer.

? How is this relevant?


> > Doing the 10 lines is very little work for you, because you now know =
what
> > it's about but for them it's much more since they have to start think=
ing about this --
> > while not even being sure whether you found a bug or have just misund=
erstood this.
>=20
> > Basically, if I get a bug report, I'd like to see
> >=20
> > 1) what the bug is
> > 2) why it is a bug
> > 3) example code so that I can *duplicate* the bug easily on my machin=
e/environment
> >    or see that my later changes have already fixed it
> > 4) exact configuration
> > 5) (optional but strongly encouraged) What this bug affects - your in=
ternal code /=20
> >    some important package &c
> > 6) (optional) potential fix
> >=20
> > Adding the java code is optional only if you cannot, for some reason,=
=20
> > make that code (i.e. you haven't found a short program that duplicate=
s
> > the problem or you are not a coder).
>=20
> Ok, in my case, the 10 lines of code would be feasible since I *can*
> code (i.e. I'm (also) a developer). So I agree this in my case.=20

Indeed.

> But in case of a regular end user/beginner programmer (as you mention
> it), it is almost always impossible to create a simple source code whic=
h
> would reveal the bug.=20

Naturally. In their case, the "source code" is instructions on how to exa=
ctly
reproduce the bug.

> Also, many end users (even in open source
> community) are not able to do a "complete" bug reports since they don't
> necessiraly know even their *exact* configuration, to say nothing of 5)
> and 6).

Yes, and as I say above, it's all "within reason". If it's not within
reason to get some info, then it can be left out.=20

The point is: if it's not a lot harder for you than for the developer the=
n
you should do it.

> Finally, there seems to be clear difference between non-open source
> community and open source community with bug reporting in general ;).
> While in non-open source community people sends only "this does not
> work" messages with *automatically generated* configuration reports, in
> open source community it is expected that (almost) all end users are
> also developers &c: when making a bug report it is assumed that you
> would provide "complete" bug report (even) containing a potential fix.
>=20
> But hey, isn't this the thing why open source software do exist ? ;)

Yes. Exactly. For closed software I've paid for it and have a right to ex=
pect
that it works! In open software I get it for free, in exchange for good b=
ehaviour
and participation.

	Tuomas



