Re: Death of Semantic Web – is it so, and how does it affect Cataloging on the Semantic web

Posting to NGC4LIB

On 12/10/2011 18:35, Sanchez, Elaine R wrote:

Does anyone on this list have an idea on what the following article means, if anything, for Cataloging on the Semantic Web, use of RDA, necessity of the granularity of bibliographic data, the future of transition from MARC to something else, difference between XML and RDF and does this affect our current projected plans? It seems to indicate that the Semantic web is dead.

I don’t usually post on this list because it is almost always above my head, but I thought this would be a good place to ask these questions. I could try Autocat, but I thought this would be a more applicable list.

Anyway, the full article is here:

Thank you so much for sharing this. I have been one of the “XML is best” guys, and that RDF is just so incredibly complicated that it can be used only with a great deal of trouble–in fact, so much trouble that I have wondered how any organization can really implement it.

The idea of “microdata”, which is much easier to implement, seems to me to be a very logical extension of Cataloging in Publication. I think catalogers could really get into it and find a real niche as they begin to understand the different ramifications of the concept “Search Engine Optimization”. (I wrote on this a bit on Autocat a few months back, where I discussed Eric Hellman’s impressive talk. See

There are lots of problems with the current version of microdata however. For a cataloger, a look at the full hierarchy that they have created is a great place if you need a laugh! Codes for “AutoPartsStore” and “ComedyEvent”???!!! It’s just too good to be true!

Still, I have thought there would be a great place for libraries in the area of extending it. The site says,

“3d. Extending
Most sites and organizations will not have a reason to extend [Hahahahahahahahahaha! – JW] However, offers the ability to specify additional properties or sub-types to existing types. 

If you are interesting in doing this, read more about the extension mechanism.” 

It can be extended, so that is critical and saves the entire project.

Naturally, microdata will be spammed by the webmasters in all kinds of ways, just as the “<meta>” area was before, with the result that the search engines had to ignore all of the information there. I think there will be great demand for library services which can supply a necessary and very desired area of trust–that is, if we play our cards right.

It would be so great if libraries could start building their own search engine that could take advantage of these tools. Maybe Hathitrust would be a good place to start… if it survives?