Question about report of Q.22/16 from last SG16 meeting (November 2006))

[이준섭]

Dear experts,

I have been made aware of a possible error in H.224 when supporting 
more than one client. Most of the implementations I know about only 
support 1 client (H.281 FECC), but there some implementations that 
also support T.140 and non-standards clients.

After consulting with the H.224 editor, we concluded that a 
typographical error was likely introduced in figure 3 in the first 
version of H.224. In the earliest complete draft of H.224 that we 
could find (dated 9 September 1993 see attached), Figure 3 is already 
present, and clearly the "number of clients" field is NOT supposed to 
be repeated.

In order to correct this, I would like to receive feedback from 
manufacturers that have implemented more than one (1) H.224 client to 
understand if implementations follow:

1) the more logical method of the attached MS-HDLL document (bracket 
does NOT include the "number of clients" field) or,

2) the published Figure 3/H.224 method (bracket includes the "number 
of clients" field).

Once I understand the backward compatibility impacts of a possible 
correction, I will submit a proposed corrigendum document to the 
upcoming meeting of SG16 (26 June-6 July).

Thank you and best regards,

Patrick Luthi
Rapporteur for Q.1/16


>  This mail has been sent from the ITU web site to itumail :
>
>Subject
>
>ITU Website feedback
>
>
>UserLanguage
>
>English
>
>MessageType
>
>Problem
>
>Name
>
>Daniele Paccaloni
>
>Subject
>
>Possible error in ITU-T H.224 (Figure 3).
>
>Comments
>
>Hello,
>
>We would like to report a possible error in H.224 which could cause 
>severe incompatibilities between terminals supporting more than one Client.
>
>The possible error is in Figure 3/H.224 (all revision up to the latest):
>We think that the \"number of clients\" byte should NOT be repeated 
>\"For each client\".
>Since the \"number of clients\" byte indicates how many entries are 
>in the following client list, we believe it should not be repeated 
>for each client entry in the list.
>
>Note that this problem may have remained undetected for years since 
>most terminals only support 1 client (H.281 FECC).
>
>Please let us know if you plan to correct this, or not (for backward 
>compatibility reasons).
>
>Thank you,
>
>.daniele

_____________________________
Patrick Luthi
Manager, Technical Standards
Rapporteur (chairman) for the ITU-T multimedia systems, terminals and
data-conferencing Experts group (Question 1/16)
TANDBERG
N-1366 Lysaker, Norway
Phone: + 47 67 125 125
Direct: +41 21 801 8420
e-mail: patrick.luthi at tandberg.com

<http://www.tandberg.com/>
http://www.tandberg.com

    
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.packetizer.com/pipermail/sg16-avd/attachments/20070604/c1e27d0c/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image0021.jpg
Type: application/octet-stream
Size: 9940 bytes
Desc: not available
URL: <https://lists.packetizer.com/pipermail/sg16-avd/attachments/20070604/c1e27d0c/attachment-0012.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MS-HDLL.DOC
Type: application/octet-stream
Size: 63425 bytes
Desc: not available
URL: <https://lists.packetizer.com/pipermail/sg16-avd/attachments/20070604/c1e27d0c/attachment-0013.obj>