ACTION-386: Review draft-barth-sniff and send comments, cc TAG

Sigh, dealing with absolutely stupid W3C HTML WG process issue has completely distracted me. Postponing.

I'm still working on this, but I'm making some progress. The main thing that needs to happen, IMHO, is to make sniffing "opt-in" on a case-by-case basis, e.g., only sniff sites you've never been to before, warn the user if you're sniffing, only sniff text/plain but not any other type, etc. etc.

The document still doesn't read that way. Part of the problem is the presumption that the sniffed type is "correct" and that sniffing is "correcting an error", when I think the language should say that sniffing is a heuristic, a guess, and in some cases, the guess is wrong and the original content-type is right, and it is the user agent's responsibility to continually make that evaluation rather than leave error-correction-of-error always the default forever.

At this point, I don't think I can get the point across by complaining, and instead I'm working on a rewrite.

I'm not going to have time to get this done before F2F without some help.

See also John Kemp's ACTION-399.

This has become a W3C/IETF Liaison issue, see:

http://lists.w3.org/Archives/Public/public-ietf-w3c/2010May/

If W3C documents make normative reference to documents in the IETF which are not actually being tracked by IETF (they're more "editor's draft", with no schedule.)


I think the last discussion on apps-discuss was around
<http://www.ietf.org/mail-archive/web/apps-discuss/current/msg01250.html>.

> It isn't an official IETF document, in that it doesn't list an intended status, nor is it being watched by an AD:
> http://datatracker.ietf.org/doc/draft-abarth-mime-sniff/
>
> I think the appropriate thing to do is to bring it up as a liaison matter; I'll do so shortly.

Sounds right to me.

There's a related HTML WG issue --
<http://www.w3.org/html/wg/tracker/issues/104> -- for which I'm supposed
to write a change proposal but didn't because of the unclear state of
the I-D.

Also, related to this, see
<http://www.w3.org/Bugs/Public/show_bug.cgi?id=9628>.

Anyway, I don't think there's any point to reviewing the -5 document again, because the review comments on -03 weren't addressed, and unless there's some procedural way of insuring that there's technical review from the affected communities, there's no point in having a TAG discussion about it.

I don't think any W3C document should make normative (or even informative) reference to [MIMESNIFF] until its at least entered into some standards track.

See first several messages of

http://lists.w3.org/Archives/Public/public-ietf-w3c/2010May/