[Sip] RE: [rohc] Requirements for signaling compression

["Price, Richard" <[e-mail]>]

Hi,

Considering a generic solution is fine provided that it doesn't delay the
availability of a SIP/SDP compression scheme. This is mainly a protocol
rather than an algorithm issue (as most of the proposed algorithms are generic
text compressors). If the protocol can also be made generic without too much
additional effort then it is a step worth taking.

Higher-layer compression might need one additional requirement - tolerance to
reordering between the compressor and decompressor. We should word this similarly
to Requirement 6, i.e:


   "The compression scheme must be able to tolerate the maximum expected message
   reordering between the compressor and decompressor."

This makes sure that the requirement is automatically fulfilled when there is no
message reordering (if per-hop compression is chosen).

Best regards,

Richard P

-----Original Message-----
From: Hans Hannu [mailto:[e-mail]]
Sent: Friday, August 03, 2001 10:00 AM
To: Price, Richard
Cc: [e-mail]; [e-mail]
Subject: Re: [rohc] Requirements for signaling compression


Hi,
Comments inline.

>These requirements seem fine to me - assuming of course that SIP
>compression is performed on a per-hop basis.


The requirements are not only considering SIP compression. The intention is
to provide requirements for a generic solution applicable for compression of
ASCII based signaling protocols, such as SIP/SDP, RTSP etc..

>What about compression between a mobile and a SIP proxy?
I believe that the requirements apply also to this case.

BR
/Hans H

>Is this beyond
>the scope of the ROHC WG?
>
>Regards,
>
>Richard P
>
>-----Original Message-----
>From: Hans Hannu [mailto:[e-mail]]
>Sent: Thursday, August 02, 2001 2:31 PM
>To: rohc
>Subject: [rohc] Requirements for signaling compression
>
>
>Hi,
>
>>From the discussions I see that compression of SIP is the major concern
all
>have, but maybe we should also look at the current version of the
>requirements document as we will discuss requirements in London, not only
>solutions.
>Chapter 3 of,
>http://www.ietf.org/internet-drafts/draft-ietf-rohc-signaling-req-assump-01
 .
>txt states the current requirements. There are most likely requirements
>missing and maybe not all are wanted...
>
>BR
>/Hans H
>
>---
>Mailing list for Robust Header Compression WG
>Archive: http://www.cdt.luth.se/rohc/
>