Federated Search Engines?

[Heather Shalhoub]

I am currently working on a committee that is reviewing federated search engines. We are looking into acquiring a federated search engine (or common user interface) for our library. We have a number of products on our radar. I’d like to know if anyone here uses a federated search service or similar product? If so, which product are you using and for how long, its advantages/disadvantages, and what the general consensus on the product is in your library: good, bad, indifferent? Any comments or observations are very welcome.

Thanks!

[Steve Watkins]

While we haven't tried implementing it locally (yet), you should definitely keep an eye on the LibraryFind project at the Oregon State University Libraries where they've been developing an impressive, open-source federated search product that appears to rival the big-name commercial products. More info is available at http://libraryfind.org/

[Bob Kosovsky]

We use WebFeat. It's nice to be able to use this tool, but the more databases you search, the slower the results are going to be. If I want to search across several databases (including our two OPACs), it can take 5-10 minutes to achieve a result. That's a long time, although I think it would take just as long with another vendor.

But I've noticed it is a tool people want to use. They just know to perform their search, go do something else, and then come back to see the hit list.

[Rhonda Gonzales]

We are also interested in implementing federated search. We have held off for three reasons. First, our ILS is not stable (we have Dynix Horizon which is going to be phased out in place of Sirsi and we're waiting to see what will happen with that). Second, we pay for some of our databases via consortia and for some of the biggest ones, our usage figures into the pricing structure. We are afraid that the federated search tool will result in more search traffic that will drive up our costs. Finally, the cost has been out of reach.

Does your library have concerns about any of the same issues?

[Heather Shalhoub]

My library is just starting the research process when it comes to FSE. We are looking at what is available and finding out what other academic libraries are using and going from there. Cost is always a concern. In the long run, my committee makes suggestions but it is up to others to make the final decisions.

[Abe Lederman]

I am the founder and CTO of Deep Web Technologies, a company which I believe has one of the best federated search product on the market today. We display results incrementally as soon as they become available and use sophisticated relevance ranking to find the most relevant results from thousands of results that a user's search might return. We have developed a very sophisticated connector language for interacting with each source being federated so that we don't "dumb down" to the lowest common denominator.

We power a number of major public sites such as Science.gov and Scitopia..org.

Please contact me -- [e-mail] to learn more about our product/service and to discuss whether we could build a pilot for your library.