![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
fleaOne of my library school assignments this week is to write a reaction to a recent strip of the "Unshelved" library comic. I chose this one:
http://www.unshelved.com/archive.aspx?strip=20070818
This recent episode of Unshelved resonated with me as [Looni] did a major stylistic and minor functional overhaul of its web site this summer, the first such revision in three years. As patrons are coming back in force in advnace of school starting next week, we are getting some reactions at the Reference Desk, especially from faculty, who notoriously hate change. (Happily I have never yet seen any apparently naked - like the patron in the strip - faculty in the [Looni] libraries; we did have a student who visited in a superhero outfit including mask a couple of years ago, but nudity is not a problem.)
The new web design (visible at library.[looni].edu) was the result of extensive user testing and library staff feedback, and includes a new logo designed by an outside firm. The original impetus for the redesign was a feeling that our old home page (available at http://web.archive.org/web/20070509052945/http://library.[looni].edu/) was too cluttered up with links, and patrons wanted more open space on the page, greater clarity, and in general something that looked more like Google (www.google.com). In short, patrons, we believed, wanted a library home page that was (mostly) "blank except for a link to the catalog and one notice." We succeeded in reducing the clutter, I think, and there are certainly now fewer links off the home page than there were (now: 38+, then: 60+). But so far the changes are mainly cosmetic (for the better, I think) and the structural revision only touches the first layer of the web site. We have a lot of work to do to de-clutter the deeper pages, and I (as a content provider and site updater, but not someone with control of many higher-level pages) am concerned that there is a lack of editorial guidance and defining structure of the site. Librarians have a tendency to run amok in creating web content and then not keeping track of it or updating it - at least, we at [Looni] do, from what I have seen in my five years here.
This Unshelved strip also speaks to a big controversy going on in the [Looni] libraries about our 'article and database' tool. We implemented MetaLib, the ExLibris product for presenting databases and allowing metasearches for articles, in 2004, and have been in the throes of a philosophical divide about it ever since (our current interface is at http://metasearch.library.[looni].edu). The argument has been back in force this summer, as we are about to implement a new public interface for MetaLib, called the X Server, that looks a heck of a lot like Google (our beta site is live at [add url]). It offers one search box for articles, which is what users say they want, and what our Systems team is pushing for. Many public service librarians say that users think they want one search box for articles, and will settle for the results they get when they use it, but never even realize that the one box can only search 10 or fewer databases (which we choose), which have to be z39.50 compliant, and the results ranking algorithm is osbcure and in some cases bizarre. Some librarians have even gone so far as to say, "Why not just recommend Google Scholar?" (something I never expected to hear in the excited period right after the introduction of Google Scholar! http://scholar.google.com/) Right now it looks like the Systems side is winning - they have data, based on click-throughs and what interfaces people are actually using - and the public service librarians will have to continue to painstakingly teach their library instruction classes that the shotgun "Quicksearch" approach may work some of the time, but for the deeper scholarly research they should be looking in subject-specific databases where they can find better, more relevant articles and have greater control over their searching capabilities.
At least nobody's retinas are being burned.
http://www.unshelved.com/archive.aspx?strip=20070818
This recent episode of Unshelved resonated with me as [Looni] did a major stylistic and minor functional overhaul of its web site this summer, the first such revision in three years. As patrons are coming back in force in advnace of school starting next week, we are getting some reactions at the Reference Desk, especially from faculty, who notoriously hate change. (Happily I have never yet seen any apparently naked - like the patron in the strip - faculty in the [Looni] libraries; we did have a student who visited in a superhero outfit including mask a couple of years ago, but nudity is not a problem.)
The new web design (visible at library.[looni].edu) was the result of extensive user testing and library staff feedback, and includes a new logo designed by an outside firm. The original impetus for the redesign was a feeling that our old home page (available at http://web.archive.org/web/20070509052945/http://library.[looni].edu/) was too cluttered up with links, and patrons wanted more open space on the page, greater clarity, and in general something that looked more like Google (www.google.com). In short, patrons, we believed, wanted a library home page that was (mostly) "blank except for a link to the catalog and one notice." We succeeded in reducing the clutter, I think, and there are certainly now fewer links off the home page than there were (now: 38+, then: 60+). But so far the changes are mainly cosmetic (for the better, I think) and the structural revision only touches the first layer of the web site. We have a lot of work to do to de-clutter the deeper pages, and I (as a content provider and site updater, but not someone with control of many higher-level pages) am concerned that there is a lack of editorial guidance and defining structure of the site. Librarians have a tendency to run amok in creating web content and then not keeping track of it or updating it - at least, we at [Looni] do, from what I have seen in my five years here.
This Unshelved strip also speaks to a big controversy going on in the [Looni] libraries about our 'article and database' tool. We implemented MetaLib, the ExLibris product for presenting databases and allowing metasearches for articles, in 2004, and have been in the throes of a philosophical divide about it ever since (our current interface is at http://metasearch.library.[looni].edu). The argument has been back in force this summer, as we are about to implement a new public interface for MetaLib, called the X Server, that looks a heck of a lot like Google (our beta site is live at [add url]). It offers one search box for articles, which is what users say they want, and what our Systems team is pushing for. Many public service librarians say that users think they want one search box for articles, and will settle for the results they get when they use it, but never even realize that the one box can only search 10 or fewer databases (which we choose), which have to be z39.50 compliant, and the results ranking algorithm is osbcure and in some cases bizarre. Some librarians have even gone so far as to say, "Why not just recommend Google Scholar?" (something I never expected to hear in the excited period right after the introduction of Google Scholar! http://scholar.google.com/) Right now it looks like the Systems side is winning - they have data, based on click-throughs and what interfaces people are actually using - and the public service librarians will have to continue to painstakingly teach their library instruction classes that the shotgun "Quicksearch" approach may work some of the time, but for the deeper scholarly research they should be looking in subject-specific databases where they can find better, more relevant articles and have greater control over their searching capabilities.
At least nobody's retinas are being burned.