December 07, 2006
OU Library Traveller - eBook Greedy
Three days away blogging doing other things (helping scope a robotics show for Techniquest in Cardiff for one thing) and it so feels like a chore coming back to it... so here's just a quick diary post to log recent OU Traveller progress before functionality gets locked down and styling spruced up ready for an official release by the Library :-)
Ok - so what's new? Looking back over the blog, the last update described a title and author lookup. Following discussions with the Library, the functionality has been tightened up somewhat and a greater focus placed on delivering OU Library services, with the result that things like a Google Books lookup are unlikely to appear in the official release.
(Just by the by, Microsoft Live Booksearch also fired up today... )
Wherever possible, libezproxifying of links to ebook resources has also been built in, which means single click-thru from the OU Traveller panel to books on O'Reilly Safari for example, as well as other e-holdings.
To produce the above screenshot, a two step lookup of the library catalogue is actually being carried out. For the particular book in question, an ISBN lookup on the catalogue returns a results page with two options - a link to the full results page for a physical copy of the book, and a link to the results page for an ebook holding. Being greedy, the script identifies the ebook link and a second background call to the library catalogue is made using this link. The ebook item results page that is consequently returned is then scoured for the link to the actual ebook resource, and it is this link that is the one that is presented in the Traveller panel.
The link text also disguises where the book is actually held. The same link text (Available online (courtesy of The Open University Library)) is also used for books held by O'Reilly Safari for example (which the OU Library also has a subscription to). What is interesting is that Safari ebook hits are not returned by the Voyager catalogue (maybe I should write a script to extend the library catalogue pages to report holdings users do have access to even if Voyager doesn't know about them? ;-)
I realised today that it's also possible to use an IP address lookup to identify whether a user is browsing from the OU network. In fact, it's actually possible to tell from the IP address whether or not the request has been made from within the OU Library building itself. Although I've not added the functionality yet, it would seem sensible to report back the actual shelf location of items held in the library via the Traveller panel for users who happen to be visiting Amazon from within the library. Or maybe not?
You may notice in the screenshot the link suggesting the user can propose the OU Library purchases the title in question. This links through to a 'book suggest' form which is also extended via the OU Traveller extension (here's an earlier example of the same approach). In particular, book author and title information, as well as publication date, are automatically pasted into the page. The No luck? link appears whether or not the book is held (in any form) by the Library, partly because it makes the coding easier, partly to promote this service. At the moment, the suggestion form is geared towards the acquisition of physical books, but it would also be handy to be able to request the purchase of electronic resources too.
Through developing the script, I'm starting to realise how the OU Library will be increasingly able to support OU students in a traditional way. Unlike most universities, the OU campus is largely 'student-free' - not surprising, given we are a distance education institution. What this has meant to date is that book lending services focus on book provision to academics and central staff, as well as loans to OU staff in the regions.
Increasingly, however, the OU Library is able to fulfill information requests using eResources, so it may be that soon we will be able to offer students a comprehensive OU Library ebook catalogue (I almost said OPAC - but they're all horrible to use, aren't they? ;-) that they can use to borrow books from instantaneously.
Another thing I'd like to see arising from the current library website redesign project is the provision of local (physical) library catalogue search services to our students.
OU students often have lending rights at their local academic library but we don't provide a single uniform location in the OU library site from which to search these catalogues (how about library.open.ac.uk/local for example, customised via a cookie or library login ID, and offering the same search box to everyone that leads to the students local library catalogue?).
The Talis Platform project is one tool we could start using, but as with using all 'free' third party services/APIs: how do we know that a) the data will be kept up to date?; and b) how do we know the service will hang around for the next few years?
Taking a baby steps approach, what I'd quite like to see is a script that will add links identified via Talis Platform to results pages like this one that in turm lead directly to the corresponding catalogue; or, even better, buttons that will let the user search the corresponding catalogue from a single search box at the top of the page...
I don't have time to pursue this at the moment, but if someone else did, I could start to use it to demo Talis Platform functionality to the library website redesign team..............
Posted by ajh59 at December 7, 2006 08:10 PM"The Talis Platform project is one tool we could start using, but as with using all 'free' third party services/APIs: how do we know that a) the data will be kept up to date?; and b) how do we know the service will hang around for the next few years?"
The data are already being kept current, in the best possible way - by those with a vested interest in keeping them that way. The Silkworm Directory (http://directory.talis.com/) part of the Talis Library Platform is open for editing and updating by *anyone* who feels inclined to do so. You're a systems librarian tweaking some arcane internal bit? Tell the Directory. You're a public library patron, whose library has just started opening on a Sunday? YOU can tell the Directory, so that everyone else benefits. Information stored in the Directory is updated instantly, and instantly available for use in your own applications via a series of web services.
Moving beyond Directory information, data on holdings is also shared with the Platform by an ever-larger number of contributors. This data powers applications such as Talis Source in the UK, and is also available via an api for consumption in other applications including Project Cenote, Firefox browser extensions for Amazon, LibraryThing, etc, and in any number of other contexts.
We have every incentive to keep the data current. The more that individual libraries come to make use of it in their day to day tasks, the more incentive *they* will have to do likewise.
We are investing heavily in the technology behind the Platform, and a wide range of applications and use cases are moving along behind this. At its heart, the Platform relies upon *our* technology and *your* data. We want the technology to succeed, and we want to deliver real benefit to you and with you. We are therefore committed to making the Platform of value to you, and to lowering the barriers to participation by contributing libraries and by competing commercial organisations.
The use case that you outline is certainly one in which services from the Talis Platform would easily play a valuable role. Richard Wallis shows something similar, calling Platform services, at http://blogs.talis.com/panlibus/archives/2006/11/amazon_plugin_g.php.
So - what can we do to help?
Posted by: Paul Miller at December 11, 2006 09:06 AM