Hello Mark,
I understand the points you are raising. First off, I support the idea of
doing this sort of work in the IETF vs competitive commercial interest
groups. It is in the Internet's best interest to have all the "competitive
parties" come together in the IETF to work on architecture and protocols
that will further the Internet's success. Right now that is not happening.
Second, there is a lot of history behind the WREC that we do not need to
rehash (i.e. the end-to-end dogma battles). It is for this reason that I
would like to see WREC finish it's work and close down. But that is my
opinion - hence a rechartering discussion should be scheduled at the next
WREC meeting.
Lastly, stacking all the various pieces of work into one WG would be too
cumbersome - bogging down the WG. IMHO - it would be better to break the
work apart into several WGs. If you look at the history of successful WGs,
the most successful WGs (i.e. things that reach draft standard status) have
been single protocol - single issue WGs. So if the
Caching/Replication/CDN/Content/ISP/Co-Location community would like to see
things reach draft standard stage, it would be in the community's best
interest to think about narrowly focused WGs working on specific
protocols/problems. If necessary, a WG just to address the architecture
could be chartered to keep all the pieces of work coordinated. Yes, there
would be push back about "too many WGs," But the community has had to do
this in the past and it can do it again.
So .... I think there should be a "CDN" BOF at the next IETF that will have
several people walk down the list of work that think could be addressed in
IETF WGs. And, WREC should meet - work on closure on the existing work - and
have a discussion on rechartering. At that point, people should start
proposing WGs to the Applications Area Directors.
Barry
-----Original Message-----
From: Mark Nottingham [mailto:mnot(_at_)akamai(_dot_)com]
Sent: Sunday, October 08, 2000 12:55 PM
To: WREC Working Group; ietf(_at_)ietf(_dot_)org
Subject: Rechartering WREC
Recently, there's been a lot of discussion in various places about the
status of WREC, particularly since there are a few other proposals for new
working groups (currently at the BoF request stage) that need to define a
relationship, or lack thereof, to WREC before they can move forward.
WREC has a somewhat difficult past, and is currently somewhat
dormant. This
may lead people to believe that the sensible thing to do is to close the
group down and split any work items off to the new groups; I'd like to
dispute this, and open more public discussion about the future of WREC.
The group's past has been difficult because it had some work items (the
taxonomy, and the known problems document) that had to be completed before
"real" work could be started. Additionally, the interception proxy issues
and misconceptions have plagued the group for some time; it's
only recently
that a clear consensus about them seems to be forming.
However, these work items are nearly finished, and interception
proxies are
an issue that can be resolved (either within the group, or in another
non-Application group). The main reason for domancy of the group
is the fact
that there have been no further milestones identified for it, so
that we're
stuck at re-chartering.
I'd argue that now is an excellent time for WREC to become an active and
useful working group; there are many potential work items for it,
including:
* content peering
* enhanced coherence mechanisms (invalidation)
* log summary formats
* surrogate role clarification
* semantic transparency issues in intermediates
* coordination with content negotiation, other groups which affect
intermediates
* proxy discovery (very important, in light of interception proxies)
More to the point, there's a real need in the IETF for a group that can
address the Web infrastructure as a whole. Highly focused groups, while
usually productive, can miss out on the bigger picture.
To this point, I'd propose that WREC re-charter as soon as
possible, with the
above work items as well as others that come to light. In particular;
* I'd like there to be open discussion with the Content Alliance
participants on the best forum for content peering. I very much
appreciate
the fact that they've opened their mailing list and documents for public
view; this is a good first step.
At first glance, it seems confusing that they are proposing a separate
working group, as content peering is squarely within the
charter of WREC.
While I can understand that this avoids some problems for them,
I think it
would be interesting to explore how their work and resources can be
integrated into WREC.
* A relationship should be established with the EPSFW effort, if
it evolves
into a WG (as it appears it may). EPSFW doesn't seem to have as much
overlap with WREC, except in that it affects proxies and
involves semantic
transparency issues. These need to be coordinated.
WREC has been languishing for a while, for a number of reasons.
This doesn't
mean that there isn't a role for it, and that it can't address new issues.
Instead of killing it and wasteing a lot of effort, let's address those
reasons and move forward.
--
Mark Nottingham, Research Scientist
Akamai Technologies (San Mateo, CA)