- From: Mike Brown <mike@skew.org>
- Date: Mon, 23 May 2005 13:45:47 -0600 (MDT)
- To: Larry Masinter <LMM@acm.org>
- CC: "'Paul Hoffman'" <phoffman@imc.org>, "'Frank Ellermann'" <nobody@xyzzy.claranet.de>, uri@w3.org
Larry Masinter wrote:
> At some point in the past, I volunteered to take on the
> "file:" draft, but I got distracted.
>
> My hope was that we could produce a document that contained
> an informational component ("what current file: URI
> interpreters do today") with a survey of current
> implementations, and some informational guidelines
> ("what future file: URI generators could do to
> be maximally compatible") and then end with some
> normative text
> ("what future file: URI interpreters should do").
>
I like all of those things, but would settle for some subset, for now,
if I had to.
> My hope would be to progress this along standards track,
> i.e., aim for DRAFT or even FULL standard. The 'protocol'
> itself meets the criteria for full standard, so we should
> be able to write a document that would stand in that
> status.
>
> I think we got into controversy because people were confusing
> what I propose as informational parts and thinking that
> they would have to be normative.
When it came up before, there were some strong feelings about whether or not
the new draft should be normative. There was disagreement as to the value of
producing an RFC that says nothing but "here's the current situation with file
URIs: FUBAR" versus going a bit further and making some recommendations for
producers and consumers of 'file' URIs.
I don't think there should be any rush to get *something* published, just so
we can put the final nail in RFC 1738's coffin. Consensus on this scheme is
hard to achieve, so it deserves a lot of deliberation and input from as many
people as possible. That's partly why I privately advocated doing the initial
draft development as a wiki, so we aren't wasting energy writing up
line-by-line comments about typos and phrasing and missing info; interested
parties would make their own changes directly, contentious edits would be
quickly reverted, and parallel discussions could occur on the list or on the
associated discussion page, as in Wikipedia. I think it would alleviate the
difficulty we're having in getting the ball rolling, and keeping editors from
tearing their hair out trying to accommodate everyone's contributions, some
of which trickle in rather late.
Speaking of which, the Wikipedia article on URIs [1] is still a work in
progress. I recently added a history section, in an effort to help
clarify the relationship between URIs and URLs, so that I could then
rewrite the URL article [2]. However, I haven't gotten around to touching
the URL article, which is currently piecemeal and quite ugly, IMHO. As
Wikipedia is increasingly becoming a reference source, it'd be great if
some more experts from this list could get on there and make whatever
changes they see fit.
-Mike
[1] http://en.wikipedia.org/wiki/Uniform_Resource_Identifier
[2] http://en.wikipedia.org/wiki/Uniform_Resource_Locator
Received on Monday, 23 May 2005 19:46:00 UTC