August 02, 2006
The Compleat Grazr/OPML Library Interface, v 0.0.1
Preamble: In So What Exactly Are Library Feeds Good For? I showed several screenshots that demonstrated how a wide variety of library related feeds could be aggregated using an OPML file and displayed within the Grazr OPML browser. In this post, I discuss this 'presentational mashup' in a little more detail, and offer it up as a entry to the Talis Library Mashuip Competition.
More and more libraries are making web feeds (RSS and Atom feeds) available covering a wide variety of Library related services including, but not limited to:
- patrons' library details (e.g. current loans and overdue books);
- lists of new books in the library;
- lists of institional open access/OAI eprints;
- lists of print and ejournal subscriptions;
- Opensearch-like feeds from library catalogues;
- subject librarian recommended websites/resources.
In addition, many content providers make book or journal information available via syndication or other XML feeds, such as:
- book contents and summaries;
- book cover graphics;
- journal current issue contents and abstracts;
- persistent search results.
Many of these feeds have been created rather 'casually', and on inspection/attempted use do not validate (e.g. using a validator such as the one found at rss.scripting.com, such is the pace
at which some of these feeds were produced. That some of the feeds remained online with errors
for so long also suggests that uptake of/subscription to them is very low indeed.
One of the "live" issues for the OU library at the current time is how best to integrate
library services within the new Moodle VLE that is being developed for course delivery from 2007.
There are several key issues when considering library service integration, not least among
them:
- What services are users likely to use?
- when and where are users likely to require these services?
- what sort of interface should e used to deliver these services?
One thing, however, is clear - if library services are consigned to their own page, they are likely to become ghettoised within the VLE and are unlikely to be used 'in context' - if at all - by the majority of users. One click away may be be one click away too many.
However, if all the library services could be collected together in an easily navigated, small embedded library panel, then they could be made available at the point of use.
The embedded Grazr OPML browser shown below demonstrates a first attempt at a) scoping out what feeds/services might be included in a feed based library interface; and b) displaying such feeds in a compact manner.
You can find a copy of the OPML file that feeds the above example at ./ousefulLibraryTest.opml. Note that this file does not contain a link to my (authenticated) user details.
The OPML file has the following structure:
...
<outline text="New Books" >
<outline type="rss" text="New books: ICT" xmlUrl="http://library2.open.ac.uk/rss/ict.xml" />
<outline type="rss" text="New books: Adult and Continuing Education" xmlUrl="http://library2.open.ac.uk/rss/adultconted.xml" />
</outline>
<outline text="Journal Current Issues">
<outline type="rss" text="The Computer Journal" xmlUrl="http://comjnl.oxfordjournals.org/rss/current.xml" />
<outline type="rss" text="IEEE Pervasive Computing" xmlUrl="http://csdl.computer.org/rss/pervasive.xml" />
</outline>
...
The <outline text=""> links define 'directories' (which may be nested, I think).
The <outline type="rss"> links define the feeds that are contained within each 'directory'.
To generate your own OPML file, save the OPML file linked to above, and then open it in a text browser. Replace the headings and feeds with your own headings and feeds, save the file and upload it to webserver. You can then view the file via http://grazr.com/config.html
As stated above, my feed bundle is just a first attempt - it does not include a section for librarian blogs, or library news and events, for example (though it could easily consume my OUseful blog feed, and maybe even an iCal (or other calendar generated) feed).
If you are a librarian out there who wants to join in the mashup competition, but perhaps don't have the programming skills required to build a 'technical' mashup, why not post some alternative OPML feed bundles that demonstrate differently structured interfaces to the library system?
Alternatively, why not show off differnt feed mixes using a webtop such as Google/ig, Netvibes, or Pageflakes?
Or come up with a library customised page on Zimbio?
All these are - I believe - acceptable things to enter into the mashuip competition...
Remember - the feeds are increasingly out there, but how many people use them. (Do you?). We are still to find effective ways for the majority of users to use them, and coming up with sensible feed mixes - and ways of consuming those mixes in context - are, I believe, still lacking...
For the purposes of the Talis "Mashup the Library" Competition, 2006 my submission is actually a two part submission: firstly, the demonstrated use of OPML to bundle library related feeds and display them in an integrated, and page embeddable, way using Grazr; secondly, the selection and organisation of feeds bundled together in the full version of the OPML file demonstrated on this page (i.e. the version of the file that includes patron loan/overdue books details (screenshots).
I would , however, encourage other people to submit 'rival' OPML files to the competition showing different ways of mixing library related feeds in a single OPML file. Being a lazy type, if someone else can find an effective mix that I can pinch, it saves me the trouble;- )
PS When I get a chance, I will post a screencast of the browser in action here...
PPS Ideally, of course, what we really need is an easy to use tool that allows users to straightforwardly create a library feed mix...
Technorati tag: MUTL06