W3C home > Mailing lists > Public > public-xg-lld@w3.org > February 2011

RE: Brainstorming: Key Issues

From: Young,Jeff (OR) <jyoung@oclc.org>
Date: Mon, 28 Feb 2011 09:47:58 -0500
Message-ID: <52E301F960B30049ADEFBCCF1CCAEF590B979F9D@OAEXCH4SERVER.oa.oclc.org>
To: "Ed Summers" <ehs@pobox.com>, <public-xg-lld@w3.org>
Cc: <public-lld@w3.org>, "Hickey,Thom" <hickey@oclc.org>
Ed,

Requests for bulk VIAF data should go through Thom. I copied him on this
reply. 

A periodic dump of the RDF (along with other representations) is made
available to VIAF contributors. This RDF dump is created as a set of XML
documents (one per VIAF "cluster") and thus the file is not XML
well-formed overall. It should be easy for us to run this through a
process to convert them to N-TRIPLE format, though, so it could be
processed directly by RDF tools.

I agree with your suggestion that sitemaps and Atom feeds could help
attract attention. These mechanisms also make it easier to gather
information in bulk, though, which presumably brings us back to the
issue of licensing.

Jeff

> -----Original Message-----
> From: public-lld-request@w3.org [mailto:public-lld-request@w3.org] On
> Behalf Of Ed Summers
> Sent: Monday, February 28, 2011 4:53 AM
> To: public-xg-lld@w3.org
> Cc: public-lld@w3.org
> Subject: Re: Brainstorming: Key Issues
> 
> On Thu, Feb 24, 2011 at 5:46 PM, Young,Jeff (OR) <jyoung@oclc.org>
> wrote:
> > To answer the question on bulk access to VIAF, the current policy is
> that
> > somebody needs to request it. So far, there hasn't been significant
> demand
> > for the RDF representation.
> 
> Where would be a good place to request it? I've had some offline
> conversations with Thom Hickey where he said there were potentially
> some licensing things to sort out before dumps could be made available
> generally. I realize it can sometimes be computationally challenging
> to dump a changing dataset like VIAF, and that there can be
> significant costs to making dumps available.
> 
> An alternative could be something like a sitemap, like what OCLC uses
> to provide access to crawlers of WorldCat [1]. This would make it
> possible for crawlers to keep synchronized views of the data. We use
> Atom in a similar way at id.loc.gov [2] since it has some facilities
> for paging, indicating when records have been deleted, and linking out
> to different representations (similar to how libraries have
> traditionally used oai-pmh).
> 
> A nice side effect of a sitemap solution is that people will find
> these troves of author information when they search for them [3]. As a
> result the VIAF URLs will hopefully get referenced more in places like
> Wikipedia, OpenLibrary, Freebase, etc. For me, this is what Library
> Linked Data is all about ... using URLs as identifiers for the things
> we care about in libraries.
> 
> //Ed
> 
> [1] http://www.worldcat.org/robots.txt
> [2] http://id.loc.gov/robots.txt
> [3] http://www.google.com/#q=site:viaf.org
> 
Received on Monday, 28 February 2011 14:49:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 28 February 2011 14:49:29 GMT