Muhammad Jaseemuddin wrote:
>  I completely agree with you that unless we have a
> signalling/flow-establishment mechanism we cannot really define the
> flow-label usage in a meaninful way. Perhaps we should wait until new
> NSIS sginalling might come-up with some usage and mechanism for this
> bit-space.

But it is the responsibility of this group to restrict their domain of
choices to things that make architectural sense. As I said in the last
note to Margaret, the participants in the QoS WGs have consistently
proven they don't understand the value of an end-to-end constant. If we
don't point out that the DSCP is their mutable field to play with so
leave the FL alone, we will end up with 2 random numbers and application
developers will continue to ignore QoS because they have no means to
express their intent that will be valid at all points in the path.

Tony


--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to