- From: David Huynh <dfhuynh@alum.mit.edu>
- Date: Tue, 19 Aug 2008 01:28:39 -0700
- To: Kingsley Idehen <kidehen@openlinksw.com>
- CC: public-lod@w3.org, semantic-web@w3c.org
Kingsley Idehen wrote:
> David,
>
> In the "nice to have" bucket, is is possible for you to use <link
> rel="dc:source" title="Data Sources" type="application/< [atom+xml] |
> [rss+xml] | [rdf+xml] >" href="<feed-information-resource-url>" /> to
> expose the list of Freebase URLs in you pages. Thus, a list of all
> URLs (once all "javascript:{}" have been replaced).
>
> Example:
> http://mqlx.com/~david/parallax/browse.html?type=%2Fgovernment%2Fus_president
>
>
> Has URLs (instead of: "javascript:{}") for each page about a
> President, but this isn't the case re. "Filter Results" DIV. Ideally,
> all the Freebase URLs could be exposed via a Feed which lists all the
> URLs (as per the <link / > suggestion above, with dc:source as the DC
> term for data sources).
>
> So one last thing, which cannot be expensive to implement since you
> are simply substituting a few more "javascript:{}" for URLs and the
> listing them is a data sources collection resource (using Atom, RSS
> 2.0, or RSS 1.0 or even OPML).
>
> When all of this is in place, I will then have multiple points from
> which to launch a SPARQL Query without the user writing a line of
> SPARQL or seeing anyting to do with RDF. They will simply feel the
> FORCE of the Linked Data :-)
It's actually not trivial to add that <link rel="dc:source" /> as
Parallax is a dynamic mostly client-side application that has only one
URL, unless you explicitly ask for permanent links. It's like Google Maps.
So, ball is in your court again. :-) There are plenty of proper web
links now to work with. Let's see some force :-) It'd be nice to have a
screencast of that, too, so it's easy to follow and forward.
David
Received on Tuesday, 19 August 2008 08:30:15 UTC