Through a talk but by David Walker from Cal State, San Marcos, he is exploring many concepts about user centered design for library systems. He is actually talking about how he has used the MetaLib ‘X-Server’ which is a set of XML APIs to develop a new interface to the MetaLib application – but many of the issues are about design, not the API.
It’s an interesting area as there has been a lot of debate about the appropriate interface to MetaLib as a federated search engine and library gateway. One of the problems we face is presenting the complexity of the MetaLib application in a way that the users will relate to. Perhaps part of the challenge is that we (as a group) are unsure how much to expose the real complexity. We perhaps have a natural inclination to tell our users how complex it all really is – while our users are clearly not very keen on understanding this.
Some nice aspects to Xerxes (the interface David has built on top of the MetaLib application) – splitting your search targets by labels such as ‘Books and Media’, then rather than pointing at specific collections, breaking down into speed of availability – ‘in our library’, ‘1-3 day deliver’ … These are some excellent ideas – we really need to learn about how our users approach their work, and start delivering interfaces which relate to this.
Some clear indications that ‘availability’ is a key issue. I have conflicting feelings about this. I can see why users want to know if something is available full-text or not – and I would myself – but I also feel that this is a severely limiting approach for serious search. It worries me (professionally – it doesn’t keep me awake at night!) that people will simply not use any references where we don’t have the electronic full-text.
Some limitations to what can be achieved by Xerxes, as not all the functionality of MetaLib is yet in the XML API – although coming this year.
Interestingly, many of the things that David is talking about would be possible in the native MetaLib interface I think. There is as much about the terminology, and translation of data from the metadata record as it is about the use of the xml api. There are some display issues that you need to have real flexibility to acheive – however, we can go along way towards it, without the extra resource needed to develop a whole new API. It is also interesting how David has harnessed the power of other applications (e.g. Google Maps) to enhance the user experience. If you know which library a book is in, why not show it on a Google Map – then you have links to directions etc as well.
Just finishing up, David is mentioning his work on ‘rss creator’. This uses the power of a federated search engine, and an openurl resolver (in this case MetaLib and SFX) to create a table of contents alerting service.
The presentation is meant to be at http://public.csusm.edu/dwalker – but not there yet. However, there are some details of the rss creator at this address.