Re: Approved TAG finding: Authoritative Metadata

On Mon, 14 Aug 2006, Karl Dubost wrote:
> > 
> > Bjoern recently posted a detailed message listing the problems with 
> > the W3C validators and the team(s) behind it, in fact.

To clarify, I was referring specifically to the fourth paragraph, where he 
highlights that the "W3C does not invest in these tools, it relies on 
volunteers", and to his detailed explanation of the lack of support from 
the working group tasked with maintaining the specifications for which the 
W3C's markup validator performs conformance checking.

   http://lists.w3.org/Archives/Public/www-validator/2006Jul/0058

I did not mean to imply that Bjoern said that the QA team was not doing 
its best given the current constrained situation it is faced with.

But all of this does not stop the fact that the validator, after over a 
year of the bug being reported, still has a bug critical to the subject at 
hand. *Even the W3C's validator does Content-Type sniffing.* If the QA 
team is responsible for the validator, then the QA team is responsible for 
this bug.

If the W3C validator itself can't get authoritative metadata handling 
correct, then why expect the vendors of other tools, let alone the authors 
of content, to get it right? If anything, this underscores why I think the 
TAG finding in question is unworkable.

I went into more detail about what I think the real problem with the 
finding is in this post:

   http://lists.w3.org/Archives/Public/www-tag/2006Aug/0030.html

...so I'll stop now.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Monday, 14 August 2006 02:38:15 UTC