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

Re: [foaf-protocols] WebID pre-alpha specification (uses RDFa)

From: Henry Story <henry.story@gmail.com>
Date: Tue, 13 Jul 2010 00:13:45 +0100
Cc: Reto Bachmann-Gmür <me@farewellutopia.com>, foaf-protocols@lists.foaf-project.org, RDFa WG <public-rdfa-wg@w3.org>
Message-Id: <565D64B3-507E-48E1-864A-B0798CF075E8@bblfish.net>
To: Toby Inkster <tai@g5n.co.uk>
There are issues with RDFa that I encountered with my work on http://webid.myxikiw.org that are worth mentioning here.

In short the main prolblem was that rdfa is fragile. In XWiki it requires the developer to rdfa enable a number of Velocity scripts. An RDFa specialist will find this quite difficult as these scripts are spread all over the place in Xwiki: Velocity scripts calling other velocity scripts recursively, can make it difficult to find out how to move up the stack.

If an RDFa specialist can have difficulty doing this, it will be way too easy for an
RDFa greenhorn to make a mistake. Having the success of foaf+ssl be so closely tied to the skills and abilities of RDFa/HTML newbies does not seem like a good idea to me.

RDF/XML is easily generated automatically, and so I think is still preferable in a world where RDF skills are still not widely available.

Therefore I think we need to find a formulation that is more neutral. I would say that official RDF syntaxes should be supported. With higher priority than other specs perhaps.


On 12 Jul 2010, at 22:30, Toby Inkster wrote:

> On Mon, 12 Jul 2010 16:19:00 +0200
> Reto Bachmann-Gmür <me@farewellutopia.com> wrote:
>> - I'm against requiring RDFA as this means I cannot expose my triple
>> store as a collection of Web-Ids without adding some templates.
> RDFa is a fairly suitable format for triple store dumps - after all,
> there's no rule saying that the HTML you generate must be pretty - a
> bunch of empty <span> elements garnished with @about, @rel, @resource,
> @property, @content, @datatype and @xml:lang ought to do the trick.
> RDFa 1.1 will be an even better format for dumps, as it removes the
> need to split the predicate and datatype URIs into CURIEs. Though
> there are some tricks for generating lazy CURIEs in 1.0, e.g.
> 	<span
> 		about="#i"
> 		xmlns:p="http://xmlns.com/foaf/0.1/name"
> 		property="p:"
> 		content="Toby Inkster"></span>
> -- 
> Toby A Inkster
> <mailto:mail@tobyinkster.co.uk>
> <http://tobyinkster.co.uk>
> _______________________________________________
> foaf-protocols mailing list
> foaf-protocols@lists.foaf-project.org
> http://lists.foaf-project.org/mailman/listinfo/foaf-protocols
Received on Monday, 12 July 2010 23:14:16 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:19:48 UTC