On 5 Jun 2009, at 18:23, Joe Touch wrote:
Lloyd Wood wrote:
On 5 Jun 2009, at 17:28, Joe Touch wrote:
Lloyd Wood wrote:
On 5 Jun 2009, at 15:54, Joe Touch wrote:
Lloyd Wood wrote:
...
Do the internals of TCP change when it's running over IPv4 vs
IPv6?
Does the interface to upper layers change?
Actually, yes - the pseudoheader over which the checksum is
computed
channges. So too does ICMP signalling, setting of the DF bit for
fragmentation discovery (clearing it doesn't have the same
effect in v6
as in v4), etc.
...so, by analogy, there's likely a need to describe HTTP over
different
transports, what is different, and what changes.
Yes - I thought that's where they were going with the SCTP stuff.
Right. But are there more general 'HTTP over transport X' rules or
conventions that need to be described?
I'd hope that a general decoupling would be useful, then a set of
mappings, one for each transport. That sort of regularity hasn't been
common - either in the IETF as a whole or in the past evolution of
HTTP,
though ;-)
Are we trying to encourage them to do that, or just overseeing how it
evolves?
Neither. We're interested in using the result, even if we have
to describe it ourselves.
http://tools.ietf.org/html/draft-wood-dtnrg-http-dtn-delivery
http://tools.ietf.org/html/draft-wood-tae-specifying-uri-transports
http://personal.ee.surrey.ac.uk/Personal/L.Wood/publications/internet-drafts/draft-wood-dtnrg-http-dtn-delivery/lloyd-wood-http-dtn-delivery-ietf-71-dtnrg-session.pdf
DTN work: http://info.ee.surrey.ac.uk/Personal/L.Wood/saratoga/
<http://info.surrey.ac.uk/Personal/L.Wood/><[email protected]>
