Comments on: Keeping subjects up to date with open data

By: jon_scott

jon_scott — Mon, 02 Aug 2010 21:23:31 +0000

I have to admit I was being a little lazy and intended the original auto-complete to simply prove the concept so I only grabbed the pref_headings. Yours is a good suggestion though and I will now go back and look at how to break out the alt_headings and include them in the UI … somehow … I can see multiple reasons why the alt_headings would also be of interest to a general user.

Cheers, I am really enjoying reading the blog and intend to keep up with everybody’s lib from now on – jon_scott

By: John Mark Ockerbloom

John Mark Ockerbloom — Mon, 02 Aug 2010 20:45:10 +0000

Thanks for the links, Jon. I wonder whether adding the alternate headings to the auto-complete module might make it even more useful. (For example, if you start typing in “gods” it could include among the suggestions either the alternate heading “Gods, Hindu” or the preferred heading “Hindu gods” pointed to by the alternate heading. Right now, it appears to just show the subjects that have preferred headings starting with what you type in.)

I haven’t looked really closely at the sort routine you suggest, but sorts are useful in the kind of updating I describe in my article. In particular, my scripts sort the obsolete headings they find by frequency of use (and can also sort by number of possible substitutions), so I can quickly find the ones that might be high priority for manual attention. Sorts can be a powerful tool in bibliographic discovery and management.

By: jon_scott

jon_scott — Sun, 01 Aug 2010 08:06:47 +0000

sort order or fitness or relevancy = approximate total of number of items held by libraries worldwide with that lcsh in the marc.650 field .

By: jon_scott

jon_scott — Sun, 01 Aug 2010 07:56:50 +0000

for your review – what are your thoughts about an lcsh worldcat discovery module ala: http://snac.wsulibs.wsu.edu/

or mesh or … ? … 14+ million other possible autocompletions: http://snac.wsulibs.wsu.edu/