[standards-jig] VoIP and Jabber

Richard Dobson richard at dobson-i.net
Tue Apr 1 15:54:19 UTC 2003

> I am disappointed by your comments Richard.

> I personally would like to see what input Craig has to offer into this
>  I have done a fair amount of work with voIP myself, and have played with
> a few ideas, nothing brought up by anyone thus far has met my requirements
> for such an implementation in jabber thus far.

I wasnt saying I was not welcoming any input, i certainly do, and im sorry
if thats how it came across I was just noting the state of play as I
currently see it and problems I see in using H.323. More and more people are
getting DSL and in a lot of those cases they might be behind a hardware DSL
router acting as a NAT, also those users may well not be savvy enough to set
it up to allow incoming H.323 (port forwards and such last I heard) which
means it wouldnt really work for those users.

> There are two levels, this I agree with your outlook... Simple initiation,
but at
> the end of the day, oob does that already, and extending oob for only
> would really be a waste of time...

Not quite, it doesnt having anything for dealing with firewalls, its just a
url for intiating a users H.323 client, not particularly integrated.

> Why not step back, look at the functionality that an h323 gatekeeper can
> take, and think about a full feature set that could be implemented via

> Things such as oob DTMF, call forwarding, conferencing etc, its all done
> a proprietary protocol stack at the moment, which is specific to the
> protocol, i.e. h323 has one, sip has a different...

> People would argue about re-inventing the wheel, but I have to disagree...
> sometimes you have to rebuild to be able to proceed past the stage you
> were in the first place... The possibilities of doing this, could open up
> whole new layer to voIP, with a nice XML based open source backend...
> well.. Everyone would agree that that is a nice thing to have.

Yes certainly I wasnt saying that we wouldnt want to do this as a jabber
protocol, Craig was promoting the use of H.323 instead and I simply stated
some problems I see in doing that. Also remember the real need at the moment
that started this off was the need to implement a voice feature similar in
operation to the voice chat in MSN and Yahoo, and its fine if we go and
extend it beyond that but IMO that is the base we need to start from, also
why is there necessarly the need to do a full VoIP implementation, arnt
proper VoIP programs more suited to the purpose rather than duplicating the


More information about the Standards mailing list