Technology Trends in Academic Libraries... and Beyond
This afternoon, I attended the ACRL WebCast on Technology Trends in Academic Libraries, presented by Roy Tennant of the California Digital Library.
He covered four broad categories in his hour presentation, which I'll try to outline using my scanty notes from the talk together with some of my own observations. At the end of this, I'm going to focus on the juxtaposition of the first and fourth trends and try to work in some real examples continuing from where Roy left off in his talk.
This is a partial outline of the four trends that Roy described.
- Better search systems (in OPACs)
- better exposure of controlled vocabulary
- better browsing opportunities
- enhanced records
- relevance ranking
- recommendations
- grouped displays
- linkages to additional content/info/services
- Roy gave the examples of
- NC State's catalog, where a search on the term "iran foreign policy" returned results that could be further narrowed by suggested topic, genre, format, era, author and other terms or browsed by subjects arranged by LC call number
- Cal State San Marcos's experimental catalog, where an interface, similar to the library's standard book browse, imposes an AJAX overlay on top of catalog searches that filtered results by items available in the library, available in 24 hours (via Cal State's consortium), or available in 5-10 days (a jargon-free description of interlibrary loan using WorldCat as the finding aid)
- Santa Monica Public Library whose innovative catalog has an extremely intuitive results interface. A search for the book The Time Traveler's Wife by Audrey Niffenegger allows users to place a hold, find more by the author, find more by topic, find items nearby on the shelf, or, most interestingly, find copies in nearby libraries using an ISBN search in Open WorldCat, centered on the Santa Monica zipcode. "A Look Inside" tab provides information such as a book summary or a review to aid the user in deciding if the book is useful.
- Roy also touched on how initiatives such as FRBR or "Functional Requirements for Bibliographic Records", which collapses individual cataloging records for a work down to a single entry which includes all instances of the work in its various expressions (such as a Russian translation), manifestations (such as its 3rd edition), and, at the library level, copies.
- Roy showed that FRBR can be a desirable display approach in such products as RLG's RedLightGreen catalog as well as OCLC's Curiouser.
- Metasearching
- search 2 or more sources simultaneously using harvested metadata stored on the metasearch server
- sometimes called "Federated Search" although that term also has the meaning of a protocol, such as z39.50, which directly queries the sources
- often includes merging and de-duping results
- often includes ability to save/email/download citations
- can include ranking strategies
- librarians want system they can search; everybody else just needs a way to find answers
- offers an interface that "de-silos" (my term) the search results
- Google has increased user expectations and impatience
- unfortunately, the examples Roy gave, Cal State's experimental Xerxes and UCLA's implementation of MetaLib aren't publically available
- Web 2.0
(Caveat lector: I'm overlaying my own spin on Roy's notes here and providing some links that were not given in the talk)- based on http/html overlayed with web services and delivered using AJAX
- web services come in two varieties
- SOAP - query sent as xml file using behind the scenes http post
- REST - query sent as url with search parameters
- response returned as an xml file
- AJAX or "Asynchronous JavaScript And XML" is the layer that asynchronously sends search requests from the user and returns the results, all without requiring that the page be refreshed in the browser. It is good for pulling small amounts of data in response to user actions.
- Google Maps is the AJAX example that is most widely known
- Mashups, according to Wikipedia, are "is a website or web application that seamlessly combines content from more than one source into an integrated experience". Roy describes them as "the opposite of silo systems" in that they allow you to mix and match content
- Roy gave the examples of Chicago Crime, which overlays crime reports on a Google Map, and iSpecies.org, which aggregates search results from a number of separate wildlife and genomic databases -- a metasearch mashup
- Microformats
- Roy described these succinctly as open data formats that are based on established standards
- Wikipedia goes into a bit more detail, describing microformats as "markup that allow[s] expression of semantics in an HTML (or XHTML) web page ... done using specific HTML attributes: class, rel, [and] rev"
- Roy's examples include hCalendar, a microformat based on the iCal standard, and COinS or "Context Objects in Space", a microformat specification for publishing OpenURL references.
I didn't take down all of Roy's concluding remarks, but two key ideas for enriching library applications were:
- open up your own library's systems so they can be used by others
- use information from other systems in your own
And this is the point where I'm going to apply a little synergy of my own.
Roy mentioned that the Santa Monica Public Library's catalog provides a link to the entry in Open WorldCat. Roy also mentioned that he used Firefox's Greasemonkey, a scripting platform that, according to Wikipedia, "allows users to install scripts that make on-the-fly changes to specific web pages". When he displayed the Open WorldCat record in his example, he pointed out that a button for his library's SFX link resolver was being displayed.
The SFX button, it turned out, was displayed via Greasemonkey scripts that place localized COinS microformats code on the the Open WorldCat page.
I mentioned an earlier incarnation of the COinS idea in this blog back in February, 2005, intrigued by a mention of the idea appeared on the Web4Lib listserv. In my first post, I referenced a page hosted by the Yale Medical Library, which is now called "COinS Browser Extensions for Your Library", but was then more awkwardly entitled "Appropriate Resolvers, Dynamically: Adding rel and title attributes to OpenURLs. A Prototype". Rather than Greasemonkey (which didn't yet exist) dynamically placing the COinS microformat on the page, it had to be hand coded onto a web page. For example, the OpenURL (NISO standard Z39.88) embedded in the Open WorldCat record has this format; a little bit complex to hand-code, but simple to pull from the cataloging record.
<span class="Z3988" title="url_ver=Z39.88-2004 &rfr_id=info:sid/worldcatlibraries.org:worldcat &rft_val_fmt=info:ofi/fmt:kev:mtx:book &rft_ref_fmt=info:ofi/fmt:xml:xsd:oai_dc &rft_ref=http://partneraccess.oclc.org/wcpa/ servlet/OUDCXML?oclcnum=52216142 &rft_id=info:oclcnum/52216142 &rft_id=urn:ISBN:1931561648 &rft_id=urn:ISSN: &rft.aulast=Niffenegger &rft.aufirst=Audrey &rft.auinitm= &rft.btitle=The+time+traveler%27s+wife+%3A+a+novel++ &rft.atitle= &rft.date= &rft.tpages= &rft.isbn=1931561648 &rft.aucorp= &rft.place=San+Francisco++CA &rft.pub=MacAdam%2FCage &rft.edition= &rft.series= &rft.genre=book"></span>
In the original scheme, the mechanism to resolve the COinS microformat into an SFX or other OpenURL button for a specific library was handled via bookmarklets, small javascripts programs that can transform the COinS microformat on the page into the desired SFX button. In a follow-up blog post, I described how I wrote a COinS bookmarklet for the SU Library.
The bookmarklet for SUL on the Yale COinS website is the one I uploaded. At the time, it seemed pretty slick despite the inherent implementation flaws. First, a user had to download the bookmarklet specific for his library. Educating him why this is a "good thing" was obviously not going to be easy. Second, the COinS microformat for the citation has to be coded into a page. At the time I first wrote about this, there were only a few places, such as the CiteULike social citation database that were embedding these. It is obviously going to be a very uphill battle to convince the mainstream databases to do the same. It is encouraging, therefore, that OCLC is taking the lead in moving the COinS microformat forward.
Athough the Greasemonkey solution is still bound by the limitations I previously described, it offers a much more integrated user experience than the bookmarklet approach in that it automatically transforms COinS into library-specific OpenURLs rather than requiring you to have to manually check for them. When I visited the Yale COinS website after Roy's teleconference, I was pleased to see that somebody coded up an SUL Greasemonkey script. Very cool. I'm going to have to try to make a list of just what sites are using COinS OpenURLs.
The last bit, before I wrap this up, has to do with Open WorldCat. Right now, unless your catalog provides a nifty "Search nearby libraries" link, you have to funnel your book search queries through Google or Yahoo. This, as everybody who has used this search can attest, is a real pain in the ass. Thankfully, much like the Firefox Search Plugin for SUMMIT that I wrote a couple of days ago, Ryan Eby wrote a Firefox plugin for searching Open WorldCat via Google. For all the details on how and why to do this, check out the LibDev blog which gives an excellent tutorial on "How to create a Firefox Search Plugin - OpenWorldCat".
So with your Open WorldCat search plugin and your Greasemonkey COinS OpenURL script installed, you are ready to begin your first trip into the exciting new Web 2.0/Library 2.0 world. It's hard to predict the final destination, but the trip is going to be a blast.
Posted by Tom on May 10, 2006
