Re: I-D ACTION:draft-whitehead-http-etag-00.txt

On Mon, 2006-03-06 at 16:08 +0100, Julian Reschke wrote:
> Julian Reschke wrote:
> > Jim Gettys wrote:
> >  > ...
> >> It would also be good to get it converted from .doc form to XML; when I
> >> looked last, the tools were not up to the job without a lot of hand work
> >> that I didn't have time for then.
> >> ...
> > 
> > An example version of RFC2616 has been available for several years now 
> 
> s/example/XML/
> 
> > (<http://greenbytes.de/tech/webdav/#rfc2616>), and I would think that 
> > probably would be a good starting point if the modest goal is to apply 
> > the collected errata, and to update references.

Said XML crashes Firefox 1.0.7.  Not a good sign.

> > 
> > If what's needed here is mainly a document *editor*, and there is indeed 
> >  consensus over here that we want that revision, then consider me a 
> > volunteer.
> > 

I think much harder to get commitment that people will actually read and
verify a draft.  *THAT* is what failed last time, and the number of
errata since 2616 were enough that you can't trust one person to get it
right.

IIRC, the draft I did covered all errata through December 2003.  I don't
know how much has transpired since then; I've not paid attention.  There
may not be a need for another draft.

Remember, if the changes are small (e.g. could be fit on a page or two
of directions to the RFC editor, particularly if they are mostly
boilerplate process changes), another draft would not need preparation,
just resubmission of the draft I prepared as an internet draft.

Whether the draft I prepared meets current process requirements for
Internet Drafts, I also don't know; any submission has to meet those
requirements.  That is part of the overhead I was referring to, that
people who are currently producing documents know and I don't, along
with going through the mail archives mail message at a time.

The RFC editor also deals with edits for references to other documents
in the usual case, so if changes since that draft are mostly in the
references, they can be done by the RFC editor.

The fundamental process here is to: 
	o submit a draft to the Internet Drafts repository, 
	o get multiple people to *READ IT*, 
	o those people state that they actually read and verified it
	  against the eratta and mailing list *in public*, 
	o discuss any remaining issues and reach rough consensus on
	  any RFC editor instructions needed, 
	o and send mail to the area director asking for action (with
	 said directions to the RFC editor attached, for any nits
	 discovered in the reads that are to be included in a final
	 document). 
If it is more than simple instructions to the RFC editor, then you get
to prepare another draft, and repeat the cycle. Wash, rinse cycle,
repeat until all the soap is out...

I believe (under process of several years ago) there is a month minimum
for individual submissions before they can be considered for action; 2
weeks for WG submissions, but we're not a WG at this date, and I'd be
amazed if you can get multiple people to read it in a month anyway, so
the minimum length of time to the point of it being in the IESG's hands
is that period from submission to when one has multiple reads completed
by credible people.

So the first issue is finding readers, in my opinion, *THEN* you can
worry about whether another draft (and editor) needs to be found.  I
don't think the editor counts as a reader, in my book anyway.
			Regards,
				Your testy editor,
				- Jim Gettys

Received on Monday, 6 March 2006 15:55:01 UTC