W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > March 2010

Re: ISSUE-1 compare and contrast Vocabularies and Profiles proposals

From: Manu Sporny <msporny@digitalbazaar.com>
Date: Thu, 04 Mar 2010 09:33:38 -0500
Message-ID: <4B8FC4C2.3040705@digitalbazaar.com>
To: Ivan Herman <ivan@w3.org>
CC: RDFa WG <public-rdfa-wg@w3.org>
On 03/04/10 04:34, Ivan Herman wrote:
> On 2010-3-4 05:22 , Manu Sporny wrote:
>> * Does not support recursive inclusion of other @vocabs
> Is this explicitly said? I am not sure this restriction is necessary
> (though it is safe, of course)

It's not said and therefore is not supported. I tried to document what
was in the current proposals (not favorable changes that we would like
to make).

You are correct, the restriction is not necessary, although I'm a bit
wary of recursively importing profile documents /AND/ using xmlns: to
declare keywords and prefixes. Seems like we're mixing in-band
keyword/prefix declarations with out-of-band RDFa processor
keyword/prefix declarations.

>> * Does not depend on @profile being re-instated in HTML5
>> * Does not address ISSUE-11
>> * Requires CORS support for Javascript implementations
> I am not sure what this means...

It means that a Javascript RDFa processor could not dereference the
@vocab document for follow-your-nose unless:

1) The vocabulary maintainer supports Cross-Origin Resource Sharing for
   the vocabulary document.
2) The Javascript implementation supports downloading Cross-Origin
   Resource Sharing documents.

> I think that, in its current format, this proposal contains keyword->uri
> mappings. Ie, no format for prefix declaration

Correct, no there is currently no format for prefix declarations.

>> Profiles Proposal[2]:
> (if my understanding is correct, Mark may correct me) This is based on
> the definition prefix declarations, ie, can be used on both; but
> requires a slight change on the way CURIE-s are interpreted in RDFa.

Correct, and we briefly discussed what those changes should be a couple
of weeks/months ago. I don't think any of us saw an issue with the
modification to CURIE processing that would recognize colon-less CURIEs.

-- manu

Manu Sporny (skype: msporny, twitter: manusporny)
President/CEO - Digital Bazaar, Inc.
blog: PaySwarming Goes Open Source
Received on Thursday, 4 March 2010 14:34:07 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:19:46 UTC