I gave a presentation for a group at NTNU Library (nominally together with my friend and colleague Joost Hegle) on Web/Library 2.0 last Thursday (2007-10-18), and outlined a few ideas about how to make our library’s web presence more user-friendly. Of course, I only provided indecipherable links on-screen, which indicates how 1.0 — or rather 0.6a — I am.
For those of you not familiar with the specific issues experienced at NTNU Library, I should add that we have a static web presence based on the Escenic publishing
oddity system. While this is probably a good solution for (newspaper) presentation, the NTNU implementation leaves an awful lot to be desired when it comes to where the library is at, and getting to where the user is. (Actually, based on my own limited experience in Escenic, I reckon that you could — given a modicum of competence in PHP and CSS — just as well use WordPress rather than Escenic; they do pretty much exactly the same thing — if you turn off comments in WordPress — just that WordPress is more stable, especially in editor mode, and costs nothing at purchase time.)
The talk began with an outline of what I am about: this can be briefly outlined as “old fashioned”. I am not a Web/Library 2.0 person, I think — following Berners-lee — that the whole 2.0 thing is jargon; I do, however, think that the changes we see in the the following respects are important:
- People use the Web more
- They use it differently
- They expect more from the Web
Given these facts, we need to focus on the following:
- More interactivity
- Web services, not applications
- Providing content that others can use, modify, etc.
- Integration of different types of services
- A tasty NTNU-specific OPAC based on feeds from BIBSYS
- A user-friendly (God forbid) version of Metalib
- Adding value
- Intelligent content for people made by people
- Phil Bradley: “Who needs Google? I need people who know what they’re talking about!”
- Intelligent content for people made by people
I also mentioned a couple of practical strategies — basically for developers — and provided good reasons for using these strategies (though not all of the good reasons). The strategies are:
- To ensure encapsulation, and thereby prevent conflicts
- To improve and speed up development
- MVC — Model, View, Controller design pattern
- Ensures that basic functionality can be developed independently of presentation, which means that the presentation layer can be changed/adapted at little cost [see references under]
- Det er godt norsk [woo!]
I also provided a couple of links, one of which is on a development machine (I don’t have access to an NTNU server that has the required software) and which is thereby not worth repeating here, but the other is available at:
This page provides access to the vast majority of NTNU library resources in a simple way: you can enter titles of books, journals and databases and return more-or-less correct results; it’s also extremely light-weight. Based on Mootools and a bit of PHP, it uses a data export from Metalib to generate autocompletions for Database titles (I will very soon add all the journal titles in our holdings as well, as well as article/book search with callback to allow checking of returned results/autocompletion).
There is also one active link to the subject page for language & linguistics, which demonstrates some more of Mootools capabilities — yes, you can disagree with the presentation — as well as some suggestions regarding functionality for such subject pages, which include:
- Links with tooltips with information about the link content
- RSS feeds: tables of contents for relevant journals, new books, etc.
- Links to other relevant resources (i.e. dictionaries, etc.)
With the exception of practical information — and, web developers, please provide this under a simple link — I don’t see much point in presenting any other content from the library, as this is largely irrelevant to the everyday requirements of users (which is finding things). At a stretch, I’d accept a ticker with links to news and other news, but this must be constantly updated, otherwise it’s not acceptable.
The web page that is not linked here was a methodology for using Ex Libris’ X-Server to make Metalib seem more like something that works. This is described to some extent in a previous posting on this blog.
The penultimate thing that I mentioned was the NTNU Toolbar, which pretty much speaks for itself.
And my final point was that subject librarians should take responsibility for their content in the various distributed systems employed by NTNU Library, rather than relying on centralized IT staff to do this job. We need quality assured resources, and the subject-specific skills required to define, describe and select them. This is where information for people from people who know what they’re talking about comes in.
Gamma, E., Helm, R., Johnson, R. & Vlissides, J. Design patterns : elements of reusable object-oriented software. Reading, Mass: Addison-Wesley [BIBSYS]
And a wikipedia article with added info, for those who aren’t totally averse to the idea!