Hi,

I have an idea for the Python-based Jabber transports. What about combining 
transports into one project? This new project will become the server side 
counterpart of multiprotocol clients like Gaim and Kopete.

How this can be done? (some ideas, not all needs to be applied (at once) of 
course):
* a name for the project: maybe PyTransports?
* a logo for the project
* a Drupal-based website
* _one_ mailing list and MUC room for all "plugins" with many people in it and 
logs/archives
* one, consistent documentation base
* similar configuration files or even having only one configuration file for 
all transports (splitted in general options and plugin specific configuration 
options)
* similar string database for translators
* using the same Python libraries
* one common SVN repository
* releasing all this as one package with a plugin for every transport

A few short-term advantages:
* coders of all transports can collaborate more easily
* translators need to translate/update only one file (less strings to 
translate)
* Jabber transports will become more userfriendly as they all will have the 
same configuration file style and/or the same library dependency (if users 
can setup one transport, they can setup all)
* less documentation is needed and nearly all can be exchanged between the 
different transports
* distribution packagers only need to create one package to have support for a 
lot transports in their distribution. Of course they always can opt to split 
it in pytransports-core, pytransports-aim,... (result: more distributions 
will add pytransports as the total userbase of the different projects 
together is bigger than the current projects on their own)
* more eyes will look to the same sourcecode
* easy to adapt protocol changes fastly in all transport plugins
-->in general: coders, users, documentation writers, webmasters, packagers, 
and translators can work together

Some long-term advantages:
* competing against multi-protocol clients like Gaim and Kopete (especially 
when a small and easy-to-install package is created that includes 
PyTransports and a small server like xmpppy that runs out-of-the-box a Jabber 
server on localhost with all transports. This opens the way for Jabber-only 
clients to have some proprietary network features that are now a monopoly of 
the multi-protocol clients).  (As I already explained on 
http://mail.jabber.org/pipermail/jadmin/2005-May/021144.html )
* more users and contributors for the PyTransports project (new ones *and* 
people who were involved in the multi-protocol client projects before)
* more people using Jabber
* PyTransports project members can ask/write JEPs when they are needed to 
bring proprietary network functionality to the Jabber world
* more transports will be created as it is more easy (new coders do not need 
to write the same amount of code and docs, to maintain a website, to announce 
new releases, to setup a mailinglist. They also easily can look to the 
existing plugins as they will act nearly the same.)
* advanced and more difficilt things can become availaible (e.g. a web 
interface for all the transport plugins)

Goals:
(1) Making the Jabber community stronger,
    (2) by accelerating the development of Jabber technologies,
        (3) by increasing the utility of Jabber-only clients,
        (4) and by copying (while improving!) features non-existant in the 
Jabber world, from the other networks.
    (5) by luring users of the competing networks to the Jabber world,
        (6) by making switching to Jabber more easy.

Competition (in the order I think the project best "attacts" them):
(1) Open source clients that connect directly to non-Jabber networks (and 
especially to proprietary ones)
(2) Same as (1), but closed source clients.
(3) The (proprietary) instant messaging networks.
(4) Transports outside this project.

A possible strategy ("Community Bodyshopping"):
(1) Make the new PyTransports project.
(2) Create a plugin for the most bad maintained or non-existing protocol 
plugin of some (or all) open source multi-protocol clients.
(3) Make this plugin much better than the multi-protocol client plugins.
(4) Eventually contribute code to the Jabber plugins from the multi-protocol 
client projects to improve their Jabber support and to make it very easy to 
use the PyTransports plugin.
(5) Promote the usage of reaching this network via Jabber/PyTransports instead 
of via the plugin in the multi-protocol clients.
(6) Target: get all users and contributors that used the multi-protocol client 
plugin before to the Jabber community (Jabber ID, PyTransports, JEP 
writing,...) so that the multi-protocol plugins for this network will become 
unmaintained and finally disappear in new versions (remark: it can be 
possible that coders from the multi-protocol project that are doing other 
protocol plugins, don't like to drop this protocol plugin. Thus they maybe 
will invest their time - that they otherwise had spent on their plugin! - to 
make necessary updates for the plugin. But this might be not enough to keep 
the whole client stable/looking professional/secure (which is a good thing 
for Jabber-only clients!).

Possible projects for such a merger:
* http://msn-transport.jabberstudio.org/ (PyMSNt)
* http://blathersource.org/ (PyAIM-t and PyICQ-t)
* http://xmpppy.sourceforge.net/ (Yahoo transport, IRC transport, xmppd.py can 
be maybe used for the mini local Jabber server idea)
* http://jmc.jabberstudio.org/ (Jabber Mail Component)
* http://jabberstudio.org/projects/jjigw/project/view.php (Jajcus' Jabber to 
IRC Gateway)

-- 
Mvg, Sander Devrieze.

xmpp:[EMAIL PROTECTED] ( http://jabber.tk/ )
From [EMAIL PROTECTED]  Sun Jul 24 20:39:37 2005
From: [EMAIL PROTECTED] (Oleg Motienko)
Date: Sun Jul 24 20:39:40 2005
Subject: [py-transports] Re: Russian characters with PyICQ-t?
In-Reply-To: <[EMAIL PROTECTED]>
References: <[EMAIL PROTECTED]>
        <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>

Hello.

> I tested latest svn version of PyICQ-t, it seems like it works well with
> vcard-encoding.patch and <encoding>windows-1251</encoding> in config.xml.

But what about of decoding icq away status messages ?
I'll test this more more precisely in couple of days.

-- 
Regards,
Oleg
From [EMAIL PROTECTED]  Sun Jul 24 21:50:46 2005
From: [EMAIL PROTECTED] (Oscar =?ISO-8859-1?Q?Hellstr=F6m?=)
Date: Sun Jul 24 21:50:42 2005
Subject: [py-transports] PyICQ Updates
In-Reply-To: <[EMAIL PROTECTED]>
References: <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>

On Fri, 2005-07-22 at 08:15 -0400, Daniel Henninger wrote:
> Folk,
> 
> I finally was able to relax and get some work done on PyICQ, so I applied
> a wealth of patches and misc other modifications.  They are in SVN now, or
> you can also grab the latest tarball from
> http://www.vorpalcloud.org/tarballs/.  These include but are not limited
> to rate limiter fixes, encoding fixes, and offline message fixes (both
> outgoing and incoming).  Many thanks to all of those who submitted
> patches!  Please test the code if you have a chance.  I'm actually
> concerned that part of it may not work with Twisted 2 very well.  (there
> was a domish.py modification and I haven't checked if Twisted 2 will
> behave the same way)  Important things to test are vcard lookups and
> messages from folk with non-english characters.

Does this mean that the issue with character encodings is supposed to be
solved? Anyhow, I will install and run the latest tarball as soon as
possible, not tonight, tired and let U know how it behaves.

> 
> Daniel
> 
-- 
Oscar Hellstr?m, [EMAIL PROTECTED]
jid: [EMAIL PROTECTED]
icq: 52604556
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : 
http://modevia.com/pipermail/py-transports/attachments/20050724/e595e964/attachment.pgp
From [EMAIL PROTECTED]  Mon Jul 25 02:49:38 2005
From: [EMAIL PROTECTED] (Daniel Henninger)
Date: Mon Jul 25 02:49:35 2005
Subject: [py-transports] Re: Russian characters with PyICQ-t?
In-Reply-To: <[EMAIL PROTECTED]>
References: <[EMAIL PROTECTED]>
        <[EMAIL PROTECTED]>
        <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>

There have been a lot of changes merged into the current SVN.  Please do
try them out with the various encoding issues people were running into
before.

Daniel

-- 
"The most addictive drug in the world is music."
     - The Lost Boyz

> Hello.
>
>> I tested latest svn version of PyICQ-t, it seems like it works well with
>> vcard-encoding.patch and <encoding>windows-1251</encoding> in
>> config.xml.
>
> But what about of decoding icq away status messages ?
> I'll test this more more precisely in couple of days.
>
> --
> Regards,
> Oleg
> _______________________________________________
> py-transports mailing list
> [email protected]
> http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports
>
>
>

Reply via email to