[Web4lib] Google Search Appliance and OPACs

[Casey Bisson]

It's a good idea, and aside from the potential for improved search  
results, the opportunity to make an entire library (all our catalog  
records + all our other web content) searchable in one place is a good  
one.

A few challenges:
+ Most OPACs are set to refuse search crawlers, or their URLs are not  
durable, or they otherwise confuse crawlers.
+ OPAC displays are not well marked up for search crawlers.
+ Google (and the appliances) do best when there's a rich web of links  
between resources[1], but very few people link to our catalogs in that  
way, so they fall back to the keyword index.

I've tested a Google Mini[2] in connection with Scriblio[3], which I  
designed to be easily indexed by search engines. One huge limitation I  
faced was that the Mini I have will index up to 100,000 pages, far  
fewer than the 330,000 bib records in my catalog[4]. But, among those  
records it did index, the results were about as good as the keyword  
indexing in Scriblio. The special Google Sauce, as it turns out, is in  
measuring links between pages. And among the pages the Mini could  
index, there just weren't any meaningful links that it could use to  
improve the findability of the 'right' resources.

On the other hand, Google.com does a fine job of indexing Scriblio for  
my library. Note the position of plymouth.edu in these searches:

http://www.google.com/search?q=a+baby+sister+for+frances
http://www.google.com/search?q=joe+monninger

It would seem that one of the most important things we can do is make  
our catalogs easily linkable and indexable, then encourage our users  
to link to the resources they value. As our users link to more  
records, their findability in any search engine should improve greatly.

--Casey

[1] http://en.wikipedia.org/wiki/Google_economy
[2] http://www.google.com/enterprise/mini/
[3] http://about.scriblio.net/about
[4] http://library.plymouth.edu/browse


On Feb 5, 2008, at 11:25 AM, Gem Stone-Logan wrote:

> Out of curiosity, has anyone experimented with using the Google Search
> appliance for retrieving information from an ILS database?  If so,  
> what was
> your experience with it?  I'm thinking of an implementation where  
> Google
> retrieves the results but then points the user to specific OPAC  
> records.