W3C home > Mailing lists > Public > public-rdf-wg@w3.org > May 2011

Re: Preparing editor's drafts -- Q's for the team contacts

From: Dan Brickley <danbri@danbri.org>
Date: Wed, 25 May 2011 16:06:00 +0300
Message-ID: <BANLkTim8X4MwjQc7r=1EmrmPpYGPuhzOhQ@mail.gmail.com>
To: Ivan Herman <ivan@w3.org>
Cc: Richard Cyganiak <richard@cyganiak.de>, RDF Working Group WG <public-rdf-wg@w3.org>
On 25 May 2011 13:58, Ivan Herman <ivan@w3.org> wrote:
> Ok, I understand. But it is not 100% clear what this would mean. From where I stand, I would opt for the creation of an HTML5+RDFa file, with a (probably off-line generated) RDF/XML and Turtle versions. This can be set up via content negotiations. This is the way
>
> http://www.w3.org/ns/rdfa
>
> has been set up. But we should realize that in view of the size of the vocabulary, this is a non-trivial amount of work.
>
> Another possibility is to mark up the RDF Schema document[1] with RDFa right from the start so that we could extract the RDF/XML or Turtle automatically by some RDFa tools (my distiller can do that without problem). There are some wizardy to be done to set up the content negotiations properly, (or we can ask the system team to put some symbolic links to our system).
>
> I am really in favour of the latter, ie, to use RDFa as part of the Schema document. If we have to have a separate document in the namespace, we are bound to introduce errors...

Yes, it has been a source of errors and confusions since RDF and RDFS
diverged in 1997 --- we get dangling pieces like rdf:value that never
got properly described. So in the RDF Core fixups the RDFS spec took
on more of a 'data dictionary' role, describing terms from both rdf:
and rdfs: vocabularies (while more explicitly delegating their formal
definitions to RDF Semantics). I think we should build on that, and
try to make some thin RDFa subset sections of the RDFS spec that can
directly appear at the namespace URIs for rdf: and rdfs:, and that can
be extracted with eg. XSLT (or Python or whatever).

We ought to make sure that we have systematic per-term HTML anchor
targets, so that eg. when linking to the Semantics spec, or the
Concepts spec, I can write a reliable link such as <a
href="semantics#term_rdf_value"> ... or #term_rdfs_subPropertyOf.
Quite likely each spec will not have something to say about each and
every class, property and other controlled terms; but whenever we do,
let's please have consistent conventions for linking (in addition to
any legacy anchor names).

Another possible tool, if we want: it is possible I think still for
http://www.w3.org/ -based URIs to be proxies for less permanent URIs,
so we  could have a proxy config directive in an .htaccess file which
reflects content from the Mercurial repo into www.w3.org URIs. Maybe
that is useful for URIs of editor's drafts? Though to be honest, there
is a syndrome where WGs often rely excessively on editor's drafts and
neglect to publish real Working Drafts. That can damage the work since
other WG's won't read these as carefully as the read a release. So I'm
fine using direct informal links into Mercurial for editorial
works-in-progress.

cheers,

Dan

ps. re Mercurial/CVS I added a tutorial link to
http://www.w3.org/2006/tools/wiki/Mercurial which I can't recommend
strongly enough: http://hginit.com/  ... the time taken to read it
will soon be repaid!

abstract, "Mercurial is a modern, open source, distributed version
control system, and a compelling upgrade from older systems like
Subversion. In this user-friendly, six-part tutorial, Joel Spolsky
teaches you the key concepts."


> Ivan
>
>
> [1] http://www.w3.org/TR/rdf-schema/
>
Received on Wednesday, 25 May 2011 13:06:30 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:25:43 GMT