W3C home > Mailing lists > Public > public-esw-thes@w3.org > November 2009

Re: [Dbpedia-discussion] Using DBpedia resources as skos:Concepts?

From: Pat Hayes <phayes@ihmc.us>
Date: Fri, 6 Nov 2009 11:59:51 -0600
Cc: Simon Spero <ses@unc.edu>, Leonard Will <L.Will@willpowerinfo.co.uk>, Alexandre Passant <alexandre.passant@deri.org>, Richard Cyganiak <richard@cyganiak.de>, dbpedia-discussion@lists.sourceforge.net, SKOS <public-esw-thes@w3.org>
Message-Id: <82D7CC9C-7430-4253-A904-3246B9E5C1BA@ihmc.us>
To: Ross Singer <rossfsinger@gmail.com>

On Nov 6, 2009, at 11:36 AM, Ross Singer wrote:

> On Fri, Nov 6, 2009 at 11:58 AM, Pat Hayes <phayes@ihmc.us> wrote:
>> Everything that is-about something is a document.
>> Everything that something is-about is a concept.
>> My problem is that this second assertion is blatantly false. I have  
>> shelves
>> full of books that are not about concepts at all. Biographies are  
>> about
>> people, not (usually) concepts of people. So at this point, SKOS  
>> simply
>> vanishes into never-never land. I have no idea what it is talking  
>> about
>> (quite literally).
> I sort of disagree here, but, at the end of the day, I'm not sure it
> matters if I coin a skos:Concept "about" the person in the biography
> on your bookshelf and you refer to the resource representing him or
> her directly.

Two points. First, you mis-state the issue (more below). But yes, it  
really does matter, not because *people* give a damn about such  
distinctions, but because the inference machinery does. Use/mention  
confusions (of which this appears to be one) are like sand in the  
gears of the semantic web, already.

Now, about the substantive point. The issue is not that there are two  
ways to refer, one more 'direct' than the other. The issue is the  
nature of things being referred to. To put it simply, are we talking  
about people (or galaxies or unicorns or...), or about library or  
thesaurus entries or records? They are all things that can be referred  
to, and the nature of the referring is the same in all cases (it is  
the 'interpretation' mapping in the RDF semantics). But people (etc.)  
are just a different kind of thing than entries and records, and have  
very different properties.

Also, for clarification. The resource does not represent anything. In  
the biography example - lets take a biography of Darwin as the example  
- the actual person, Charles Darwin, is the resource in question, that  
the biography is about. The name "Charles Darwin" (a string of 14  
characters, including the space) refers to the man, who is the  
resource. You cannot link to Charles Darwin, only to data/records/ 
representations/descriptions of Darwin.

> Eventually, we'll figure out some way to link those two
> together and most people probably won't care that I think that the
> subject heading for the Grand Canyon
> (http://id.loc.gov/authorities/sh85056381) is somehow describing some
> abstract "aboutness" of http://sws.geonames.org/6619202/ and not just
> the physical location.

But inference engines trying to make sense of your data together with  
other data will care. Or rather, they will break if you get this all  
muddled up.
> What I'm saying is that a Philosophy 101 debate is not going to derail
> the semantic web.

This is not a philosophy 101 debate, it is an ontological debate. And  
ontologies are the very stuff of the semantic web.

Pat Hayes

>  If I think there's a Platonic ideal of William
> Shakespeare that Elizabethan scholars are actually writing about and
> you don't agree with that but the rest of my data and assertions are
> valuable enough to you, you'll figure out how to work it into your
> application and your worldview.
> -Ross.

IHMC                                     (850)434 8903 or (650)494 3973
40 South Alcaniz St.           (850)202 4416   office
Pensacola                            (850)202 4440   fax
FL 32502                              (850)291 0667   mobile
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
Received on Friday, 6 November 2009 18:01:01 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:45:59 UTC