[...] MetaLib (with SFX) is where the bells and whistles are really obvious. There are varying degrees of increased complexity in the interface depending on how deep you go. This is great for the librarians, but unlikely that students would use it. In fact Casey at MaisonBisson, also in attendance with me, states that “only 0.0067% (YES, less than a hundredth of a percent!) of the searches on our OPAC get “limited” to specific languages, locations, dates, or material types” in his article The High Cost Of Metasearch For Libraries. The interface seems to be minimally customizable, limited to headers, footers, and CSSS. However, there are “real” APIs for MetaLib, a separate product they call XServer. It is well documented and seems to be just the right amount of useful, but more cumbersome than a mere REST-like interface. [...]

What do you think of this article?

http://www.adtmag.com/article.asp?id=11385

Unfortunately, vendors don’t sell to patrons, but to librarians. They have to add the full Boolean bells and whistles to get past the all-librarian screening committee. Most are senior staff, raised on Dialog.

[...] A9, the search engine from Amazon.com, does some pretty interesting things that libraries should be aware of. First, any library considering a metasearch product should look at what can be done for free, and second, libraries should take a look at the OpenSearch technology that drives it. [...]

[...] What about RSS, XML, OpenSearch? WordPress solves the RSS feed for us (look at this URL to see). A feature-complete XML API, is a bit further off, but maybe somebody wants to pitch in to help solve that one? And full OpenSearch support, taking advantage of the suggested and alternate search features, is my next big project (here’s where I’m going with that). [...]

Our library system is using a mix of tools: MetaFind from Innovative Interfaces as the “connector”, but running our own front-end (in PHP) that queries other databases thru MetaFind. Our relevance rank tweaked Metafind’s “relevance ranking” (which was, “10 results from whatever database answered first, 10 results from the second, etc) to ignoring the response time of each database (but preserving each database´s original “relevance” rank. However, it turns out that most users didn´t like the relevance algorithm for each individual database… it didn’t seem google-like enough. Brittanica ignores phrase matching as “higher relevance”, ProQuest seems to put articles higher based solely on term occurence… so taking whatever (limited) results came back from each metasearch, we now implemented an algorightm with high scent (pushing results up based on term occurence, phrase occurence, and newer publication date) which seems to be working better (I guess we need more studies to find the right scoring weights). I know the vendors won’t get to this soon enough, so why not tweak our metasearch engines like this in the meantime? BTW, if you want to look at this algorithm tweak, look at our results (sorry, it’s in spanish!)…. Link to bad automatic translation. =)

[tags]metasearch, relevance[/tags]

[...] Now, the question for libraries is when are we going to demand OpenSearch interfaces from our information providers? The inclusion of OpenSearch in IE7 more than gives it critical mass, but so far it seems to be just something a few progressive library-types are experimenting with. In the short term, imagine how improved our metasearch tools would be if based on fully-implemented OpenSearch feeds (with the facets and suggestions). In the long term, I can’t imagine any aspect of a library’s online services not touched by this technology. [...]

FYI:

http://libraryfind.org/

LibraryFind is…

an open source metasearch application developed by librarians for libraries, built with Ruby on Rails.

Some Current Features:

Built-in OpenURL resolver
2-click find workflow
Ability to locally index collections
Web-based administration
3-tiered caching system (to improve speed of searches)
Customizable user interface

LibraryFind® is software developed by the Oregon State University Libraries, funded in part by a grant from the State Library.