RE: CfC: Publish HTML5 Microdata as First Public Working Draft and a new HTML5 Working Draft


> Although Working Draft publications do not require consensus, I  
> encourage you to consider withdrawing this objection for the following  
> reasons ...


I don't withdraw my objections recorded in:

which apply to this CfC.

> - The HTML WG charter specifically calls for us to include extension  
> mechanisms: "The HTML WG is encouraged to provide a mechanism to  
> permit independently developed vocabularies ... to be mixed into HTML  
> documents." Microdata is one way to mix in such vocabularies.

But the actually charter which you only selectively quoted says:

 "The HTML WG is encouraged to provide a mechanism to permit
  independently developed vocabularies such as Internationalization 
  Tag Set (ITS), Ruby, and RDFa to be mixed into HTML documents."

Microdata does not provide a means of including Internationalized
Tag Set, Ruby, OR RDFa, and so "Microdata is one way to mix in such
vocabularies" is false, and your misquoting inappropriate.

> - There is precedent for this Working Group to publish a separate  
> Working Draft that adds an extension mechanism to HTML, namely HTML 
> +RDFa. No one objected to that publication as out of scope, either  
> before or after publication.

I didn't formally object at the time, but I didn't think
it was a good idea, and said so in my poll comments.
"No one objected to that publication as out of scope" is false.

In any case, the precedent cited should not continue to be


> Working Draft publications do not have to have  
> consensus and do not even meet requirements. In the spirit of wider  
> review and openness to exploratory development, this Working Group has  
> previously approved Working Drafts that not all members agreed with on  
> a technical level.

The Microdata material has already been available and
discussed, and so does not qualify for "spirit of
wider review and openness to exploratory development".

In any case, we're too late in the process of the W3C HTML 
working group to publish, at this time, under this charter,
new material that was not part of HTML4, and for which there
is not a clear preponderance of opinion that the technology
covered by the document is, or should be, part of HTML.

Even if there were reasons in the past for allowing Working 
Draft publications that do not have consensus or meet
requirements, this is not part of the current documented 
HTML Working Group Decision Policy and should not be.


Finally, I sent in privately a comment from a colleague
and metadata expert, which I believe reads on the decision
to publish Microdata in HTML WG.

Although working group members may prefer to hear opinions
first-hand, I think the formal W3C process allows 
representation of opinions by others (although this is
not clear in the HTML Working Group Decision Policy.)

For the record, what I sent was:

--- Original Message ---
From: Larry Masinter 
Sent: Sunday, December 20, 2009 10:38 PM
To: Paul Cotton; Maciej Stachowiak; Sam Ruby

Subject: Additional objection to "keep microdata"
A colleague sent me a comment, but wasn't able to get
into the HTML working group in time to make the 12/17 
deadline for comments. I thought I would forward them
anyway. The comment was in the context of reasons why
neither Microdata nor RDFa may be "the answer" and should
be split into separate drafts.

Since you're collecting opinions, and these points
don't seem to be made in any of the other postings,
I thought I would forward these comments.

I think there are 2 aspects to this.

The first is that I agree that unconstrained vocabulary extension is  
very important, and that use of namespaces is a good way to get this.  
I think the Microdata proponents are excessively blinded by fear, or  
maybe by concern for lowest end use cases (the talk about namespaces  
getting lost in copy-and-paste). Real web development needs tools,  
tools can check namespaces. Low-ish end use will quickly learn to keep  
the xmlns attributes close enough. Lowest end raw text editor use will  
screw up lots of things besides metadata namespaces.

The second is that the distributed nature of both RDFa and Microdata  
introduce practical issues with metadata management. It will take more  
work to find and process all of the little bits of metadata. This is  
not a deal breaker, it is a practical matter of work. The bottom line  
might be viewed as a fundamental question of why an HTML designer  
wants RDFa instead of RDF - what is wrong with RDF, what advantages  
does RDFa have?

Received on Saturday, 9 January 2010 19:58:11 UTC