Re: ISSUE-76: Need feedback on splitting Microdata into separate specification

On Thu, Oct 15, 2009 at 1:26 PM, Shelley Powers
<shelleyp@burningbird.net> wrote:
> "taken out of the spec eventually..." This one bothers me. There's an
> assumption, and I've seen it before, that the same group that exists now
> will somehow "own" HTML throughout the ages. That Ian Hickson will continue
> to be the sole editor of HTML, through HTML6, and HTML7, and so on.
>
> That is an exceptionally bad attitude to have, and inherently destructive to
> HTML5.
>
> We should take the time to do HTML5 right. It may need modification later
> on. It may need change. But a goal should be to create a specification that
> can live on, as long as possible. That it is built in such a way that it
> continues into the future. It should have only what it needs.

I never said otherwise.  Those who support microdata think it will be
successful and should always remain in the spec.  *If* they're wrong,
though, it's not like it spells the end for RDFa or other superior
technologies.  This is no different from any other feature.  Some will
inevitably be failures.

> Is that your defense of Microdata? Leave it in, because we can just pull in
> a later version of HTML? I can't believe that, you must have more extensive
> reasons for your interest. Could you provide those?

I don't care whether Microdata or RDFa end up in the spec, not in the
spec, or ditched entirely.  I do think that it would be more
productive for the HTMLWG to resolve it one way or the other, however.
 That way it can move on to discussing other issues where discussion
might actually prove fruitful.

> Winner? The only winners here should be the people who end up with whatever
> comes out of this working group.  Our effort shouldn't be a contest. It
> should be an example of cooperation.

That's a nice ideal, but sometimes it doesn't work, and some form of
arbitration is necessary.  We reached that point a long time ago
regarding RDFa vs. Microdata.  The issue has already been arbitrated
in the WHATWG by the editor, in favor of Microdata.  In the W3C
further recourse is available to the ones who lost in the WHATWG, so
take it and let's get this over with for good.  If you can work out
something that satisfies everyone, good luck, but I think there's
little hope of that at this point.

> In the mean time we're in the middle of a discussion. Do you have specific
> reasons why you personally need and want Microdata, and why you feel that
> Microdata _must_ be a part of the HTML5 specification?

No.  I never said I think any of those things, and I don't.
Personally, I don't see why it matters either way what spec anything
is in.  Microdata is specced by the HTMLWG, RDFa is specced by the
HTMLWG.  Anyone can feel free to use either.  Splitting Microdata out
to a separate spec now (as opposed to three years from now if it's
unsuccessful) isn't important enough to be worth arguing over.  Which
is all the more reason to just resolve it one way or another.  If the
RDFa advocates would drop this, or the Microdata advocates agree to
it, without needing to invoke a formal decision-making procedure, then
all the better.

On Thu, Oct 15, 2009 at 2:41 PM, Shelley Powers
<shelleyp@burningbird.net> wrote:
> Let them both have their chance in the marketplace. If both survive, fine.
> If only one survives, then perhaps at some future time, it can be
> incorporated into HTML5. Or not.

You could apply the same argument to any feature with a dubious
future.  There are a lot of things in HTML5 that don't have any
implementations yet, and not all of those will get enough
implementations to justify staying in the spec.  That's why HTML5 is a
Working Draft.  It's not likely to become a Recommendation for at
least ten years, so if Microdata is a failure, it will be possible to
split it out any time before then.  No rush; the HTML Working Group
has more important things to think about, that actually affect what
user agents have to implement in the near future.

Received on Thursday, 15 October 2009 19:00:22 UTC