RE: Revisiting namespaceDocument-8

 
Norman Walsh wrote:

> At the recent f2f, the TAG discussed namespaceDocument-8 
> again and considered how we might make progress. I've published 
> an outline of our ideas at
>
>  http://www.w3.org/2001/tag/2005/06/23-rddl/
>
>with some examples. The central idea is to get past the roadblock 
> of picking a single format by employing GRDDL.
>
>Comments, please.

Firstly I think it would be a fine idea to add

<head profile="http://www.w3.org/2003/g/data-view">
  <link rel="transformation" href="---.xsl"/>
</head>

To RDDL (1 and/or 2) as a mechanism by which a transformation to RDF is
specified.

GRDDL alone does not satisfy our original aim in RDDL of both:

1) a human readable format
2) a machine readable format

In that nowhere is it ensured that the namespace document is human readable.
Assuming that the namespace document is authored in some type of (X)HTML
that is transformable into RDF we are essentially at a variant, at the very
least, of one of the RDDLs which leads us back to either RDDL1 or RDDL2.

Personally I don't have much preference for RDDL1 over RDDL2 or RDDL2 over
RDDL1. The advantage of RDDL1 is that it:

1) is much more deployed than RDDL1 
2) uses Xlink (for those who see this as an advantage)

Whereas the advantage of RDDL2 is that it is a "cleaner" format.

When I tried to point http://www.rddl.org/ directly at RDDL2 I received an
onslaught of complaints from people who are using RDDL1 and strongly
objected to my swapping the document format out from under them (I saw the
complaints as entirely reasonable and gave RDDL2 a new URI i.e.
http://www.rddl.org/rddl2.html )

So although GRDDL certainly has much merit, it doesn't *itself* solve the
namespaceDocument-8 problem although hopefully RDDL+GRDDL is an acceptable
consensus.

Jonathan

Received on Tuesday, 28 June 2005 21:28:04 UTC