From Christian Bizer et al., DBpedia--A Crystallization Point for the Web of Data, _Web Semantics_ 7 (September 2009), DOI:10.1016/j.websem.2009.07.002
"Wikipedia Categories. DBpedia contains a SKOS representation of the Wikipedia category system. The category system consists of 415,000 categories. The main advantage of the category system is that it is collaboratively extended and kept up-to-date by thousands of Wikipedia editors. A disadvantage is that categories do not form a proper topical hierarchy, as there are cycles in the category system and as categories often only represent a rather loose relatedness between articles."
I thought the statement of advantages and disadvantages was succinct and expressed well the inevitable trade-offs in such a system. I think there's an inherent tension in the Semantic Web between these, and it's something we'll just have to learn to live with.
Ed
-----Original Message-----
From: Next generation catalogs for libraries [mailto:NGC4LIB_at_LISTSERV.ND.EDU] On Behalf Of Weinheimer Jim
Sent: Wednesday, November 04, 2009 8:02 AM
To: NGC4LIB_at_LISTSERV.ND.EDU
Subject: Re: [NGC4LIB] Tim Berners-Lee on the Semantic Web
I made a mistake. There is a namespace for "Prisoners of war" but it is in some kind of category (still don't understand it all yet)
https://subj3ct.com/subject?si=http%3A%2F%2Fdbpedia.org%2Fresource%2FCategory%3APrisoners_of_war
Jim
James L. Weinheimer j.weinheimer_at_aur.edu
Director of Library and Information Services
The American University of Rome
Rome, Italy
________________________________________
From: Next generation catalogs for libraries [NGC4LIB_at_LISTSERV.ND.EDU] On Behalf Of Weinheimer Jim [j.weinheimer_at_AUR.EDU]
Sent: Wednesday, November 04, 2009 4:33 PM
To: NGC4LIB_at_LISTSERV.ND.EDU
Subject: Re: [NGC4LIB] Tim Berners-Lee on the Semantic Web
Ed Jones wrote:
<snip>
I don't expect us to throw the baby out with the bathwater, and indeed we've been playing with LCSH ever since records containing it moved online. Tinkering with LCSH seems the most likely future, and the tinkerers have my best wishes. Maybe we will get it to work "good enough". But this is not a desirable future. It's simply that we no longer have the resources to design and build a real alternative for records in an online world. Structures are cheap, content is expensive.
</snip>
That last part is well put, but I have a different slant on it. We don't have the resources to do as you say--no one does. But "folksonomies" (I hate that word!) are coming on strong. They are popular and people find them really useful. I think a glance at sites such as subj3ct.com and dbpedia reveals how our expertise is needed for management, "improvements," and so on.
When people make their own "folksonomies," they are essentially doing the same things we are: organizing resources by different concepts and plopping labels on them. It seems to me that we could cooperate to make our materials more accessible to them and their materials more accessible to us. Since it has been shown that we are "trusted voices," probably these people would accept our "input" so long as we do not end up trying to take "control."
And this brings us back to the main purpose of this thread, about what Tim Berners-Lee said. Different communities and even different individuals will build their own structures, there's no getting out around that and it should be seen as a positive thing. It's the content they need. That goes for full-text materials as well as metadata.
As an example, perhaps this page:
http://dbpedia.org/page/List_of_prisoners_of_war
Could utilize these searches somehow:
Prisoners of war --United States --Biography.
Prisoners of war --Germany --Biography.
Prisoners of war --Japan --Biography.
and vice-versa. In fact, there is currently no page:
http://dbpedia.org/page/prisoners_of_war
but there is the one for *Lists of* prisoners of war.
Do we have anything to contribute? Certainly, and a lot!
James L. Weinheimer j.weinheimer_at_aur.edu
Director of Library and Information Services
The American University of Rome
Rome, Italy
Received on Wed Nov 04 2009 - 12:39:43 EST