“Repository Ecology” is a term we’re using for some new work in the repository domain and while it is difficult to explain what exactly Repository Ecology I thought I would start by explaining why I like the metaphor.
The work is being undertaken by colleagues in the Repository Research Team, and is described on their wiki. I suppose that once you have information landscapes and an information environment, the idea of ecology as describing how different types of thing interact is a natural progression. Well a repository ecology is, I guess, a part of information ecology, and in fact there’s some history to the idea of Information Ecologies, with books by Nardi and O’Day and Davenport being published a few years ago. (A related term term Information Ecosystem crops up in AI research, but I think that is something slightly different.) It’s also interesting to see that Dave Cormier and others have been thinking about ecologies too. Anyway, here are my reasons why I like the metaphor, for which I’ve been heavily influenced by chapter 4 of Nardi and O’Day, which is available online.
Ecology is an approach to studying a complex mess of interacting things each with its own aims (or no aims), each mutually dependent on the things with which it interacts. I think this far better describes the field that we want to study than some other approaches. For example, an architecture can provide you with a plan of something that you want to build, but it’s often not a good approach to describing what you actually have. This is quoted in Davenport:
Since we never fully implement our models, we never have a useful map of the structure and location of our current information.
An ecological study can be broad and sweeping covering a vast area, like the Russian Steppes or Amazon rain forest, or focus right down on a single pond or tree in great detail, while still recognizing that the system being studied is connected to other systems that are not being studied but are still important. I think the idea of choosing a level of granularity for a study while not isolating what you study from the wider system is a strong one.
In ecology the things being studied are divided into species, which seems to match with the approach of considering different types of service and roles of player in an information environment. One approach to ecology is to look at the role of one species in an ecosystem, and the role it plays in interacting with other species (e.g. do any other species depend on the way species being studied distributes resources). I think this is one way in which our repository ecology study can be seen (though perhaps repositories represent a genus rather than a species, since there are so many different types of repository each with their own characteristics).
Two species-related concepts from ecology seem to resonate with studying information environments: firstly a healthy ecosystem requires a diversity of species; secondly some species are more important than others (e.g. bears in distributing nutrients around forest environments) even when they are not particularly abundant, these are known as keystone species.
Ecology recognizes that the systems it studies change: populations vary number and age-profile, species adapt and evolve, the system as a whole can react (positively or negatively) to the introduction of a new species or other change in environment. Our funders are keen that we remember that we should study repository ecology from the point of view of gardeners (or maybe farmers, but hopefully sustainable, organic ones) trying to produce something, rather than as conservationists trying to maintain the status quo.
Location is important in ecology, with concepts such as habitat and niche. This is useful when gathering information from people who interact with information systems, since most informants will be able to tell you about what happens in the place they occupy in the big scheme of things (running an institutional repository, using and creating learning resources, and so on) without needing to take a view on the bigger picture.
Also on the theme of gathering information about an information environment from the players in it, I think it might help that “ecology” has rather positive connotations for most people. Think about the alternative approach of going to someone and saying that you would like to talk to them about their role in “The System”, even worse “The New System” we are going to build for you to work in. Well, such an approach wouldn’t encourage me to cooperate 🙂 .
OK, there’s a few thoughts on why I like the ecology metaphor, the work being done by my colleagues in the repository Research Team is looking at how to map it to information systems: what exactly is the equivalent of a species or a habitat? How do you describe their interactions? We also need to decide whether what we’re doing really follows on from the previous Information Ecology work or are we picking up the metaphor afresh. We’re going to try out a couple of case studies applying it (one at this year’s JISC CETIS conference in November). If you’re interested in more of the previous work we’re looking at, we’re tagging resources on our del.cio.us account as InformationEcology.