Link This |
Email this |
Blog This |
Comments (3)
LCSH Again
April 2, 2008
Having only blogged less than a week ago about the site where you can Browse LCSH a year or so after it was announced, I'm determined not to be caught flat-footed again. So this is likely the first time you've heard about this new LCSH site from the Library of Congress.
Ed Summers, a library technologist at LC and a long-time Code4Lib participant, has launched http://lcsh.info/ which provides unique access to the Library of Congress Subject Headings. What he did was to take the LCSH dataset harvested by Simon Spero (see my earlier post for more information on this) and parse it into SKOS -- an effort led by the W3C as part of its Semantic Web effort.
I found it interesting that Ed used the Simon Spero data rather than the real data, which one presumes he would have access to at LC, but knowing the difficulty of working within a large organization I can understand that it might have been easier to use the unofficial data set (freely available on the web) rather than going through the machinations that might be required to get the official data. That being said, Ed has support in this work from no less of a personage than Barbara Tillett herself, who encouraged him in this work and helped him get involved with the W3C efforts.
All of this is apparently a lead-up to doing something on the web with the official data, live and updated. This will also give Ed time to go beyond what he terms a "naive conversion" since he hasn't yet created any custom RDF vocabularies for representing the details found in the data. He also intends to make the data searchable, probably with Solr. This will make the data much more useable by both humans and machine processes searching for appropriate headings.
More information on the project can be found on Ed's web site. Specifically, there are some web slides on the project.
I am very happy to see this project happening, and I encourage the Library of Congress to press forward with full support. Making the live, updated information available in machine-readable form for next generation catalog projects could provide the means for some very interesting experiments. The fact that this very topic is being discussed at this time on the Next Generation Catalogs list NGC4Lib is telling. There will be a growing demand for these kinds of machine services, which we certainly recognize where I work, and I'm hoping that Library of Congress administration does as well.
Posted by Roy Tennant on April 2, 2008 | Comments (3)
