Paul,
Yes, you are correct. H.323v3 requires at least H.245v5.
You are correct that these paragraphs appear to be contradictory. The intent was that H.323v3 systems use H.245v5 or higher. But since you forced me to look at the text, I can see that we also agreed that H.323v2 systems may use H.245v3 or higher, as well. (This was introduced as part of the Implementers Guide in May 1999 and is reflected in the H.323v3 text.)
I agree that the confusion between the two paragraphs needs to be addressed, however.
Paul
----- Original Message ----- From: "Paul Long" plong@SMITHMICRO.COM To: ITU-SG16@mailbag.cps.intel.com Sent: Sunday, February 27, 2000 11:40 AM Subject: Re: H.323 Annex D
Paul,
H.323v3 requires H.245v5, not H.245v3.
Summary/H.323v3: "Version 3 products can be identified by H.225.0 messages containing a protocolIdentifier = {itu-t (0) recommendation (0) h (8) 2250 version (0) 3} and H.245 messages containing a protocolIdentifier = {itu-t (0) recommendation (0) h (8) 245 version (0) 5}." and Summary/H.323v3: "Support of H.225.0 (1999) and H.245 (1999 or later) as identified in messages above, shall be the singular requirement and definition of H.323 systems which are H.323 Version 3 compliant."
Interesting... Note that the second passage conflicts with the first in
that
the second says that H.245v5 "or later" shall be used but the first
passage
says that H.245v5 _only_ shall be used with H.323v3. Which is it? IMO,
we'll
run into even more interoperability problems by decoupling H.245 versions from H.323 versions. Also, H.323v2 (1998) is Decided, so it cannot be modified to allow other versions of H.245.
Paul Long Smith Micro Software, Inc.
-----Original Message----- From: Paul E. Jones [mailto:paul.jones@TIES.ITU.INT] Sent: Sunday, February 27, 2000 12:47 AM To: ITU-SG16@MAILBAG.INTEL.COM Subject: Re: H.323 Annex D
Francois,
By default, I believe that editor would be me. This annex was intended to work with H.323v2 systems and later, but you are correct that there was no way to specify T.38 until H.245v4.
Of course, H.323v2 systems used H.245v3. H.323v3 systems used H.245v3 or higher (the key element). So, I suppose the question is: is H.323v3 required? Surely not, as there are implementations out there. We have argued about whether H.323v2 systems could use later versions of H.245. The outcome was "no"-- perhaps that needs to be revisited.
There have been recent changes to H.245 ASN.1 to support T.38. At least there have been some proposed changes. Unfortunately, I missed much of that discussion. The meeting report says that those changes were added to H.245v6.
Mike, is it true that the T.38-related changes from H.245v4 to H.245v6 were nothing more than additions to H.245-- not "changes", per se. Also, did the proposed changes (from TD8/WP2, I believe) make it into H.245v6?
As for Annex D, I am considering just pulling that into the H.323 document, rather than keeping it as a separately published annex-- does anybody object to that? That would certainly address questions related to references going forward. However, we need to reach a decision about the current publication. I suspect we need something for the IG, but the problem is that H.323v2 is "old news" now. What to do....
Paul