Re: Record of http discussions at the Paris WWW conference

> As a process issue, there's little (no?) justification for the ground work
> laid down by the subgroups over many weeks/months to be tossed in favor of
> last minute changes by the editorial group.

Hmmmm, process .... this is Larry's domain, but here is my take on the
subject.

Nothing in the draft 03 document is sacred -- anything can be *removed*
if there is a good reason for it (or just a lack of rough consensus).

Anything in draft 03 which was not subject to open review by the WG
is subject to complete rewrites -- that is just a fact of life.
That includes about 80% of caching, persistent connections, and
negotiation, because the subgroups failed to introduce a complete
set of differences to the WG when they were supposed to do so, and thus
we are left with the situation of having to make last-minute rewrites
to the document itself.

It is too late to introduce new protocol elements unless they allow us
to simplify other parts of the protocol that haven't been implemented yet.

> Alternates, opaque vs. transparent, content negotiation structure.  I'm not
> willing to see it fall out of the spec without significantly more convincing
> reasons other than a hallway discussion wanted to shrink the spec at the
> last minute.

If there is argument over something like variant-ids, then the default
action will be to remove variant-ids (not, as you say, keep them in the
spec.) unless the arguments were already considered (in public) by the
caching subgroup and consensus obtained there -- in this case, the answer
is no.  Even with consensus, anything in the draft which is questioned
will have to be defended if it is to remain in the final draft.


 ...Roy T. Fielding
    Department of Information & Computer Science    (fielding@ics.uci.edu)
    University of California, Irvine, CA 92717-3425    fax:+1(714)824-4056
    http://www.ics.uci.edu/~fielding/

Received on Wednesday, 15 May 1996 17:53:50 UTC