RE: [Sip] INFO considered harmful

[Orit Levin <orit@radvision.com>]

I (also) vote for the direction #2 from the draft (i.e. convert INFO into a
proper framework).

I don't like using neither SIP Events nor INFO as general mechanisms for
implementing applications only because they happened to be collocated with
an existing SIP stack.

I do think that both mechanisms are very useful for features closely related
to the SIP call control (once these features are well-defined and
standardized). An alternative would be maintaining an expensive separate
transport channel for each of the light and complementary to the call
control applications...

On a different note, I would like to stress out that in general events vs.
requests (when properly applied) would be different in semantics. In other
words, it doesn't make sense for certain applications to use events and vs.
versa. Each mechanism stands on its own and both are required. 

Orit Levin
Chief Architect
RADVISION
Phone: +1.201.6896330
Video: +1.201.6896430


> -----Original Message-----
> From: Dean Willis [mailto:dean.willis@softarmor.com]
> Sent: Saturday, December 28, 2002 1:29 AM
> To: Jonathan Rosenberg
> Cc: sip@ietf.org
> Subject: Re: [Sip] INFO considered harmful
> 
> 
> 
> Jonathan Rosenberg wrote:
> > Folks,
> >
> > For some time, I've been complaining about the continued abuse of the
> > INFO method for things that should be done other ways (frequently, not
> > using sip at all). I've written up a summary of the problems and a
> > proposed path forward. Specifically, I'd like to obsolete INFO and
> > replace it with a spec that is approved ONLY for SIP-T.
> >
> > Until the I-D appears, you can pick up a version at:
> >
> > http://www.jdrosen.net/papers/draft-rosenberg-sip-info-harmful-00.txt
> > http://www.jdrosen.net/papers/draft-rosenberg-sip-info-harmful-00.html
> 
> Jonathan and I have had "extended discussions" on this. I'm more
> inclined to think that INFO needs specifications for specific usages,
> much as we use "events packages" for SIP Events (SUBSCRIBE, NOTIFY). In
> my own completely non-original way, I call these "INFO Packages".
> 
> I think that the SIP-T usage of INFO is a perfect example of an INFO
> package. However, I think it is NOT the only reasonable use for INFO.
> Unsolicited in-dialog "stuff" is pretty darned useful. All we really
> need is for the usages to be differentiable, negotiable, and
> well-documented. Admittedly, INFO doesn't have any of those three
> characteristics right now, but I think we can probably fix it without
> breaking too much existing stuff.
> 
> Anybody else have a third opinion?
> 
> --
> Dean
> Not Speaking From The Chair.
> 
> _______________________________________________
> Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
> This list is for NEW development of the core SIP Protocol
> Use sip-implementors@cs.columbia.edu for questions on current sip
> Use sipping@ietf.org for new developments on the application of sip
_______________________________________________
Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use sip-implementors@cs.columbia.edu for questions on current sip
Use sipping@ietf.org for new developments on the application of sip