Before launching in I should reveal my perspectives and biases. With our doctoral proseminar at GSLIS (UIUC) we have just been reading Swanson’s Undiscovered Public Knowledge and Latour’s Science in Action, so these have been fresh in my mind. I consider myself a Computer Scientist interested in interface design and collaborative working and learning.
Firstly, do I even agree with the premise? Is Information Science actually a Science? Or is it more like Engineering? Or is it more like Social Science. Or is it more like part of the Humanities?
I could give evidence for all these. My gut reaction (background, training) is to see many similarities with the engineering approach. Thus I’m interested in the pragmatics. Take the bright shiny toys of computing and figure out if they can be put to use solving much older problems. In the process I may end up asking more fundamental questions, and I may start asking computer science to work on different kinds of things to help me in my work. Thus I would see IS in many ways paralleling of a lot of Victorian technologies which powered ahead with the pragmatics of getting things to work at all (iron bridges, steam engines etc.) and subsequently inspired scientists to explore more fundamental aspects (that is my understanding of the history which could be wrong). Thus, to take a poor example I’d be less interested in engaging the interest of a Boyle (gas laws) than an Otto (built first automobile) and more interested in a Ford (created a low cost mass market product). Bad example because Boyle predated Otto (I think) and I’d like an example where the pragmatists got a rough version working that then piqued the interest of the scientists.
So my Problems are challenging, but may not be so much fundamental as pragmatic. I really like Marcia’s listing:
How do we build systems (computer systems) and create or alter social organisations (e.g. libraries, workgroups, etc.) for them to operate in?
How does the information system co-evolve with its work environment? How does the use of an information system as integrated in people’s lives CO-evolve with the functionalities available in that system? Simple example – if you have card indexes for author and subject but not title, you aren’t likely to do much title searching. When you first move to an OPAC you are likely to begin using it in the same way that you used the familiar card catalogue. But eventually your searching may alter to take advantage of the new features, so you might end up doing more title searching because you find retrieval of an imperfectly remembered title is more effective than retrieval of an imperfectly remembered author name.
Similarly, how do the requirements for improved functionalities CO-evolve with available existing functionalities?
What are the qualitative changes in experience and support technologies as we move up through orders of magnitude of ‘stuff’. Let’s say books are our arbitrary unit of measure. In what way is a library of 1000 books qualitatively different from one of 100 thousand, 10 million, 1 billion, 100 billion? My hunch is it is to do with greater black-boxing and other ways of coping with immensity by greater abstraction. The same thing applies at the smaller level of chapters, research papers or even Swanson’s facts lurking unindexed within documents but potentially in combination of great value.
Quite reasonably we have been concerned about access and information poverty. But how do people cope with the complexity of so much stuff, of suddenly becoming information wealthy? We see clues in people making sudden transitions, such as undergraduates attempting term papers in a research university’s library, or a dedicated overseas graduate student moving to the USA from an environment where she had read everything on the topic she could get hold of, and slowly coming to the realisation that ‘read everything relevant’ is impossible.
How do people cope with learning how to use new information technologies? The constantly updated versions of retrieval engines, new web-based OPACs, new databases and their fancy front ends, etc. These systems are means to ends, so how much effort do you bother to put into learning them? How much effort do you put into learning them efficiently? If you don’t have information on the cost-benefit trade-offs of learning it is entirely rational not to bother (despite how galling this may be to librarians and computer scientists).
How do we factor the costs of getting information into information use and decisions about trying to get that information? I believe we need at least informal qualitative measures of a total information cost function, so that we can argue how a new functionality would alter the parameters of that function, and talk about the benefits of obtaining information in terms of cost-benefit tradeoffs By total information cost function I mean including becoming aware of an information need, deciding to do something about it, searching, reading, thinking and incorporating that new knowledge back into one’s work. I think we need such a total cost measure to help in justifying the opportunity cost of that total activity compared to ignoring the ‘information itch’ and carrying on with what you were going to do otherwise. (We normally assume that the latter course is sub-optimal – but by how much?)
How do we design systems that are ‘good enough’ for a vastly expanded population of users as well as high powered systems for the skilled user, as well as ways to help a person transition from one to the other? This is something that many people are already working on.
How can we put people back into our information systems? Special pleading – this is my hobby horse. I would claim that a library, including a digital library, without other people is a sterile environment. Not only boring but not conducive to growth and the generation of new ideas. I’m claiming that people cope with the complexity of stuff and the complexity of learning all the interfaces to all the stuff by interaction with other people, and so we need to build people back into our systems. The undiscovered public knowledge thing of Swanson is also amenable to chatting with others and their volunteering nuggets of information that correlate.
Finally, can we get clues for functionalities for better systems by
creating a Gold Standard? Something we can aim in the direction of even
if we all acknowledge it is in practice unattainable. One way may be to
assemble a Dream Team of librarians and make them available to an information
worker (could be a professor, a general practice physician, a paralegal,
whoever). The Dream Team are available to try and answer any question using
all the resources of a major University. We could then try and measure
the results and the impact of those results. Then by looking at the articulation
of the problems by the knowledge worker and the actions of the Dream Team
we could get insights into functionalities that could provide aspects of
what was done at vast expense for that knowledge worker for many more people
at far lower cost.