W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > June 2011

[Draft] RDFWA WG comment on the HTTP Vocabulary in RDF 1.0 document: consider changing the 'http' prefix

From: Ivan Herman <ivan@w3.org>
Date: Thu, 16 Jun 2011 11:46:38 +0200
To: W3C RDFWA WG <public-rdfa-wg@w3.org>
Message-Id: <73E7906D-6843-47CD-AD29-9A99DD85DC3B@w3.org>
Dear all,

thanks Shane for having drawn my attention on a small issue around the HTTP vocabulary, namely the fact that they still use the 'http' prefix. I composed an official comment of this Working Group for their Last Call, but I would prefer to get an approval from the WG before sending it out. The LC comment period is over for a week but, hopefully, they would still accept the comment.

Actually... it may be better if somebody else sent the comment, simply because I will be off-line for a while starting this coming Sunday. Manu? Shane?

In any case, we should decide on that today.




Dear ERT WG,

apologies to be late on this Last Call comment, we hope that you would still take this into account. The (editorial) comment is on the HTTP Vocabulary in RDF 1.0 document[1].

As you may know, the RDFa 1.1 specification (see [2] for its latest version) provides a very flexible ways for the usage of URI-s and Compact URI-s. Essentially, if an RDFa attribute is used like this:

<.... rel="pref:ref" ...>

then the processing algorithm:

- decides whether a URI has been associated with 'pref', in which case the final URI is the concatenation of that URI with 'ref'
- otherwise, the pref:ref is considered to be a URI, and that will be used for in the final, generated RDF

This approach has a caveat: there may be issues if somebody defines a prefix 'pref' when 'pref' is also a define URI scheme, and RDFa authors are advised not to do that. (Actually, some RDFa tools may also raise warnings if this happens.)

Another information is that RDFa 1.1 defines the notion of an RDFa default profile; essentially, a number of prefixes are pre-defined for any RDFa 1.1 file. The exact content of the prefixes is still under consideration (see [3]) but it is fairly well accepted that any RDF vocabulary, defined as a W3C Recommendation (or note) will end up, eventually, in this list. In particular, the early, pointer methods, the content, and the HTTP vocabularies will fall under this rule when finalized.

However, the current HTTP Vocabulary document uses 'http' as its prefix in all the examples; the same prefix is also used in the EARL Schema document[4]. For the reasons outlined above, the choice of this prefix would be a mistake for RDFa, hence the plan to use the 'ht' prefix[3] in the RDFa default profile. In our view, it would be much better if the same prefix was adopted in the original specification documents, too; we believe, regardless of the RDFa issues, that it is not a good practice to reuse the name of a well known URI scheme for vocabulary abbreviation.

On a similar, though less important issue: there has been some discussions with the Dublin Core Metadata Initiative (DCMI) on the prefix that should be adopted for the dcterms vocabulary (see the thread starting at [5]). The final advise we got from DCMI is to use the prefix 'dc' for http://purl.org/dc/terms/ in the default RDFa profile[6]. You may consider using it, too, for the sake of consistency, but that is clearly at the discretion of the Working Group.


Ivan Herman
In the name of the RDFWA Working Group  

[1] http://www.w3.org/TR/2011/WD-HTTP-in-RDF10-20110510/
[2] http://www.w3.org/TR/2011/WD-rdfa-core-20110331/
[3] http://www.w3.org/profile/rdfa-1.1
[4] http://www.w3.org/TR/2011/WD-EARL10-Schema-20110510/
[5] http://lists.w3.org/Archives/Public/public-rdfa-wg/2011Jun/0015.html
[6] http://lists.w3.org/Archives/Public/public-rdfa-wg/2011Jun/0037.html

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf
Received on Thursday, 16 June 2011 09:43:40 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:19:52 UTC