W3C home > Mailing lists > Public > www-rdf-dspace@w3.org > May 2004

Longwell using W3C TR data [was Re: Simile Deliverables review; apr 27 deadline]

From: Ryan Lee <ryanlee@w3.org>
Date: Tue, 04 May 2004 16:15:50 -0400
Message-ID: <4097F9F6.4080403@w3.org>
To: "Butler, Mark" <mark-h.butler@hp.com>
Cc: SIMILE public list <www-rdf-dspace@w3.org>

Butler, Mark wrote:
>>http://www.w3.org/2004/04/simile/browse/index.html
> 
> The only thing is my original intention with Longwell was to definitely not
> have any naked URIs in the UI! I think they are ok for RDF savvy folks, but
> not for the rest of the world.

I had hoped to get things running with no URIs in the display before
letting the demo get out into the wild, but that didn't quite happen.

> So I would be very keen to try to improve how the data is rendered. Some of
> this can be done by rules processing - for example one potentially useful
> rule would be if a document is a version of another resource, then that
> documents inherits the title of that resource so for example
> http://www.w3.org/1999/08/WD-SVG-19990812 would get the title Scalable
> Vector Graphics (SVG) 1.0 Specification. However unfortunately not all
> abstract specs have titles, so that wouldn't solve everything. Another rule
> would propagate titles the other way, e.g. there is one version of
> http://www.w3.org/TR/rdf-concepts/ that does have a title, so if an object
> does not have a title, and another object is a version of it and does have
> the title, then add the title to the first object. 

I did some rule processing on a portion of the data with cwm, including 
your titling suggestion.

> From the user point of view, I think there are some more useful possible
> facets e.g. year (this would involve some syntatic processing to extract the
> information from date). Contributor and author would also be useful, but
> Longwell has a difficulty dealing with bNodes at the moment which I think we
> need to fix.

There's now a prominent publication year facet.  I'm working on the 
bNode issue in general; hopefully the author and working group 
information can be used as facets in place of activity statement 
citations.  Perhaps a rule for assigning a label to the report short URI 
would be in order for the 'version of' facet as well.

-- 
Ryan Lee                 ryanlee@w3.org
W3C Research Engineer    +1.617.253.5327
http://web.mit.edu/simile/www/
Received on Tuesday, 4 May 2004 16:15:55 EDT

This archive was generated by hypermail pre-2.1.9 : Tuesday, 4 May 2004 16:15:57 EDT