W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > March 2013

Re: HTML+RDFa 1.1 PR Transition Paperwork

From: Manu Sporny <msporny@digitalbazaar.com>
Date: Wed, 13 Mar 2013 00:01:18 -0400
Message-ID: <513FFA0E.3090500@digitalbazaar.com>
To: RDFa WG <public-rdfa-wg@w3.org>
On 03/12/2013 01:53 AM, Ivan Herman wrote:
> I think, in any case, that we will have to contact Thomas and Ralph
> informally with the solution we are proposing before we issue a
> formal transition request. 


> - I can see your point on the <link>/<meta> issue.

Good. :)

> - I think we still have other references we make into the HTML5
> document (algorithms) that should be cross-checked for stability.

I'll make another pass tomorrow, but I'm fairly certain I already did
this. That is, we can't call out the entire HTML5 parsing/re-parenting
algorithm because it is gigantic, complex, and would make the whole spec
optional. It has multiple stable browser implementations, and is as
stable as we can expect anything in HTML5 to be.

> - However, I do not think the RDFa 1.1 dependence issue will fly. The
> process is fairly clear on that issue: a Rec cannot normatively
> reference a document that is not a Rec itself. 

I don't think it's useful to point to the process document in this
instance. I think it's more useful to point to why the process document
says what it says: to ensure implementation stability. I think either
approach achieves the intended protection that the W3C process is
attempting to achieve (stability). I also think that W3C has already
done something like this before HTML5 (OWL2), and that rule is the one
that needs to be generalized to all specs. HTML5 isn't really special in
that respect... but we could spend weeks arguing the finer points of this.

I've taken your advice in the latest spec and just marked the features
as "non-normative" with an explanation in the SoTD about when the
features are expected to become normative. I think this is what you
wanted. The wording is a bit rough right now, so any suggestions on how
to clean it up would be appreciated:


To view a diff-marked version do: CTRL-ALT-SHIFT-S and click 'diffmark'

> P.S. We will probably give more details on the implementation part,
> because we are actually requesting jumping over a step. But that is a
> detail.


-- manu

Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Aaron Swartz, PaySwarm, and Academic Journals
Received on Wednesday, 13 March 2013 04:01:49 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:19:58 UTC