[jdev] TINS implementation in PSI and Jabberd2 started

[Jens Mikkelsen]

Hi Jacob,

Looks exciting. For you information I currently working on a simular
project at DIKU (Institute of computer science Copenhagen). The project
title is VoIP in Jabber. Lots of danish guys working on this. ;)
The project is also proof of concept, but more focused on NAT and
firewall traversal and p2p. Maybe we can share ideas.

If you are new to XMPP you might want to read my bachelor project from
last year on encryption in Jabber. Theres a pretty good description of
the Jabber protocols in it. I'll send it to you if you like.

Regards 
Jens Mikkelsen

On Mon, 2005-04-11 at 12:05, Jacob Bunk Nielsen wrote:
> Hi
> 
> Just to let everyone know.
> 
> My friend Sebastien Yerly and I are working on implementing TINS
> (JEP-0111) in PSI and jabberd2 as a project at our local university.
> The implementations will not be production stable implementations, but
> proof of concept. We hope to release any code we write for others to
> use, if they should wish to do so.
> 
> OK, jabberd2 already supports TINS as it complies with the last part
> of section 2.4 of RFC 3921, that says it must pass on anything it
> doesn't understand. However, we are also writing a proof-of-concept
> transport for SIP. In addition to performing voice chat between PSI
> users, our plan is to be capable of calling a regular SIP phone from
> within PSI before we finish.
> 
> We have investigated other options for voice chat and transporting
> voice to other networks, and have decided that TINS is the best way
> for reasons We're not prepared to discuss here. You are of course free
> to agree or disagree with this as you please, but commenting on it
> will not change our decision, so please do not flood the list with
> comments on this. It has already been discussed on the
> standards-jig-list.
> 
> Our goal with the project is to show that it is possible to make a
> good integration of instant messaging and voice chat with open
> protocols. We believe we can do this by showing that TINS will work,
> and that it is easy to handle and relatively easy to implement.
> 
> We currently ignore the whole problem about NAT/firewall-traversal,
> which is of course also quite essential to it's usability in the long
> run, but not relevant to what we want to show at the moment. Since it
> is possible to solve the NAT/firewall-problem when working with
> IP-telephony based on SIP/SDP/RTP (e.g. by use of a STUN-server) we
> believe it is also solvable with our approach if one should wish to do
> so.
> 
> In the course of our project we might come up with additional examples
> to put into JEP-0111, as it urgently needs this. Whether we have time
> to write some nice examples or not will probably be decided by the
> amount of time we have left over by the end of our project period.
> 
> We are both relative new to the concept of XMPP, but as we are to hand
> something in by the end of July we hope to have something ready a
> little before then ;-)
> 
> Should you want to contact us directly, we are reachable both by XMPP
> and email. Sebastien can be reached at s031039 at student.dtu.dk and by
> XMPP at seb at jabbernet.dk. Jacob can be reached by email at
> jacob at bunk.cc and by XMPP at jacobbunk at jabbernet.dk
-------------- 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: <https://www.jabber.org/jdev/attachments/20050412/b87bc8a5/attachment-0002.pgp>