W3C home > Mailing lists > Public > public-egov-ig@w3.org > March 2010

Re: Technical issues impacting government use of linked data

From: Gannon Dick <gannon_dick@yahoo.com>
Date: Sat, 13 Mar 2010 14:50:39 -0800 (PST)
Message-ID: <317634.48507.qm@web112620.mail.gq1.yahoo.com>
To: public-egov-ig@w3.org
Cc: Josh Tauberer <tauberer@govtrack.us>, Cory Casanave <cory-c@modeldriven.com>
> 1) That given a data URI, there is no standard way to programmatically
> access the metadata about the resource.

The Linked Data 'movement' started, as I recall, in the use of http: 
URIs as the convention for all RDF resources....

IMHO, the biggest impediment to linked data standardization is not the link, or the data, it's the perception of redundancy for authors in one Document Object Model or another. 

The owl:sameAs is not the same logical assertion as owl:equivalentClass.  Take for example a data base and a HTML page.  Both presumably have some sort of URI, which points to the root of the DOM at a fixed point (not the "title").  You can give a database a "title", so that's ok.  But you have to understand that the data base is all /html/body or to put it another way, that /html/head/@profile defaults to <ANY>|(null).  What you really want the profile to default to is <RDF:rdf> or some rdfs:subClassOf (Dublin Code, SKOS, OWL, FOAF etc.).  Only then will the meta data be "exposed" (as opposed to "printed") properly.  In the same way as you added a "title" to a data base you could add a fixed "profile" + rows of elements (Collection).  For linked data to "work", either a retrofit of data bases (with title and profile) or a retrofit of HTML (either profile or /html/head/meta/@xlink:type="locatorLink") is necessary. 

Received on Saturday, 13 March 2010 22:51:12 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:00:43 UTC