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

Ian Hickson, Tue, 12 Jan 2010 00:32:22 +0000 (UTC):
> On Mon, 11 Jan 2010, Tab Atkins Jr. wrote:
>> On Mon, Jan 11, 2010 at 6:17 PM, Ian Hickson <ian@hixie.ch> wrote:
>>> On Mon, 11 Jan 2010, Maciej Stachowiak wrote:
>>>> 
>>>> Alternately: Ian, would you be OK with changing the title from "HTML5 
>>>> Microdata" to either "HTML+Microdata" or "HTML Microdata" or 
>>>> something similar?
>>> 
>>> Not really... that would give people the impression that Microdata 
>>> wasn't part of HTML5, which I believe it should be.
>> 
>> As opposed to giving the impression that it's part of HTML?
> 
> Having just "HTML" in the name makes it sound like it's an independent 
> spec that one can consider part of HTML, or ignore. It makes it sound like 
> a candidate for "relevant specification", in HTML5 terms. I believe 
> Microdata should be considered an integral part of HTML5. 

Comment/Question:

> Whether that is 
> by having a single specification for HTML5, or having HTML5 split into 
> modules with Microdata being one of them, I don't really mind. I would be 
> fine with calling the draft "HTML Microdata" or just "Microdata", provided 
> that the spec clearly stated it was part of an HTML5 family of 
> specifications.

HTML+RDFa says: "This specification is an extension to the HTML5 
language." Given the WG's decision, Microdata and HTML+RDFa should say 
the same thing - whatever they say - about the relation to the HTML5 
language.

> What I object to is making Microdata a second-class 
> citizen that, e.g., validators can validly claim is not part of HTML5.

There you touch upon a subject w.r.t. to versioning which has been 
largely ignored by this group. Currently, if we do 

	http://validator.w3.org/check?uri=http%3A%2F%2Fwhatwg.org

then the validator auto-recognizes the validation profile between 18 
available HTML and XHTML profiles.

You have put in the HTML5 spec that HTML5 can simply be extended 
through writing a spec which extends it. However, you have not placed 
inside HTML5 a mechanism by which a validator can know which version of 
HTML5 to validate against.

If this WG continues to refuse to make a choice between HTML+RDFa and 
Microdata, and if it also do not place a versioning mechanism in the 
non-extended HTML5 language, then I cannot imagine that 
validator.w3.org will forever operate with a HTML5 validation profile 
which (like validator.nu already does) includes Microdata, but doesn't 
include HTML+RDFa. And, at any rate, without a versioning mechanism, 
then merely the presence of Microdata and HTML+RDFa means that there 
will be more than one HTML5 validation level profile. 

We can EITHER fully include both HTML+RDFa as part of the HTML5 
language. OR we can start to operate with a versioning mechanism. OR we 
can require that validators, whenever they detect something which is 
not part of "HTML5 proper", that they perform a heuristic check and try 
to validate against another HTML5 profile.
-- 
leif halvard silli

Received on Tuesday, 12 January 2010 15:59:24 UTC