I have been using recombinance quite a lot in the last year or two to talk about how network flow affects structures. You can tell that I am generationally challenged: I should be saying remixing, which is cropping up in places in the same sense. This is happening to structures at all levels. Here is a compressed and reductive list:
- We are used to seeing metadata flow between repositories: libraries developed an early mechanism to share records between nodes and have an extensive bibliographic apparatus. More recently, OAI-PMH has provided a mechanism for publishing processable metadata for consumption by recombining services. We have also seen the emergence of RDF and XML schema, namespaces, and application profiles which extend this recombinant ability to ‘data elements‘.
- We have always assembled collections of content objects, and seen them flow between collections. These flows are facilitated in a network environment, and resources may appear in a variety of aggregations. Again, Content has become less ‘solid’ also: think of what has happened with music. We can recombine tracks in various ways, on various devices, in various collections, for various purposes. The same is happening with other constructs, where we want to decompose and recombine differently. Taking images or figures from journal articles or chapters from text-books, for example, and recombining and contextualising them in courseware is an ambition. Related to this, we have seen the emergence of ‘content packaging‘ standards – METS, SCORM, MPEG DIDL – to bind particular resource combinations.
- With network services, recombination/remixing is very clear. There is a clear move towards more fine-grained services which can be recombined to meet application needs more flexibly. In fact, flexible service recombination is the goal of the service oriented approach. One reason that this is important for libraries, is because they want to be able to mix library services into the e-learning mix, or into the campus portal mix, as well as into the library mix. Related to service orientation are on-demand services and outsourcing, the move to source particular business processes from third parties.
- And this moves us to the organizational level, where increasingly we see that organizations themselves may articulate organizational components of different types to achieve particular ends. Think of an organization like Cisco, for example. In the library world, libraries already rely on a range of cooperative and other groups, of various sizes and scopes. My own view is that libraries will increasingly want to secure shared services, which will mean relying on other organizations, whether it is OhioLink, OCLC, JISC, the California Digital Library, DEF, Ithaka, ….
What is happening is that at all these levels structures are internalizing the network, and are adapting accordingly. At all levels the need for complex interactions is driving interest in better facilitating decomposition of structures and flexible recombination to meet service needs. To package various materials into a learning object. To deliver various library ‘channels’ into the enterprise portal. To add various databases to the metasearch. And this points to the importance of interoperability. Interoperability is the ‘recombinant ability’ an object has, the ease with which it can be remixed in different combinations to create value. To caricature a long term resident of the British Museum Reading Room: all that is solid melts into flows.
