Areas of change in repositories and digital libraries

What are the key changes going on in the world of digital libraries, repositories, and academic life? Over the summer I’ve been thinking a little on some of this - in part been prompted by my aforementioned change in job and in part because I’ve been on holiday and it’s afforded a few moments of reflection. I think these are a number of areas where current practice is beginning to change or be challenged. They certainly aren’t all new, but for one reason or another they’ve popped up on my radar over the past few months.

This is neither an exhaustive list, nor authoritative, but I’ve put some thought into what I think is changing how we manage, find, and use digital stuff and may as well share it. Time permitting I will turn some of these into fuller posts which point to some clear examples and have a bit more discussion, but as a first pass…

1. What, Who, Where, When, Why?
   • Metadata is changing and so must our perceptions and tools. We’ve had non-bibliographic metadata for a while; but increasing tools and users want and can use other forms of metadata. “What is this about and who is it by?” are still key questions but “What is this for?”, “What do you think of this?” “Is this still the same object as it was three years ago?” “What has been done to this object since it’s been in the system?” “Where is this?” “Who owns this?” are all now questions being asked of digital resources. Tools and users are looking for as much information as they can get their hands on. Sometimes this information stands in the object’s stead (like the ‘traditional’ catalogue), but increasingly it’s information that sits alongside the object and provides connections (to other objects or for tools).
2. Universities (and other organisations)  and are beginning to have the opportunity to join up the different systems and data they hold.
   • Conversations about data duplication are not new and often the problems besetting institutional connectivity are cultural rather than technical, but there does seem to be some sense change.  Technical change and new connections may not always be on the scale of Cardiff’s engagement with SOA, but maybe linking student or staff registration services with library/ e-resource user management is a little closer, and when the technical systems are in place perhaps we’ll get better at data management.
3. As more institutions begin to consider what they do with their scientific datasets, digital curation is becoming a real challenge/ practice.
   • Datasets are unforgiving and as more institutions begin to consider how they will manage the data they produce and as more funders expect good data management better practice is being forced onto the agenda. For example UK Research Funders’ Policies for the Management of Information Outputs. Interestingly enough at what might be considered the other end of the spectrum there has also been some JISC-funded work into the curation of learning materials (which are often regarded as ephemeral).
   • Reuse of datasets as more institutions develop VREs may increase the demand for/ on local dataset storage
4. Resource-oriented views of the world are growing in prominence.
   • this has been discussed extensively elsewhere but there does seem to be a trend to making repository and digital library resources more web-friendly. This shift is most evidenced in OAI-ORE.
5. Annotation is here to stay
   • We’re getting used to annotations whether we use tools built into systems or external tools. When we chose between 10 relevant items what Joe in Newcastle thought about the article, image, or food processor does influence our behaviour, and that of the users of our services. More importantly we need to make sure that it’s easy for users to ‘annotate’ our stuff in external services - repositories and digital libraries need add this article to my del.icio.us or digg buttons as much as any other online resource.
6. Mainstream web protocols are seeing more use in the library community
   • more and more repositories now come with refinable RSS feeds for new items.  Atom is seeing a lot of use as part of a deposit API (SWORD) or as a binding for OAI-ORE. These are some examples of mainstream web protocols increasingly used by libraries. This means the protocol development doesn’t depend on community development or maintenance and that, alongside the specific library tools, there’s a greater chance you can take what you produce and mash it into whatever new tool appears in the wider web community. [The importance of this may lurk behind some of ORE's current discussions about to implement ORE-resource maps in ATOM. ]
7. CRIG motto “the coolest thing to do with you data will be thought of by someone else”
   • People are going to want to metadata and resources you create or manage in ways you don’t anticipate and, if you let them,  some of what they think of will be good. Of course not everything should be freely reusable and not everything people do with it is going to be good but, in the context of publicly-funded instituions freeing up metadata and content should be our default starting point.
8. Institutional-related content is not going to only be held in the institutions
   • Stuff connected to institutions is increasingly going to be all over the place -  academics deposit in subject repositories or slideshare and content produced about institutional life ends up on blogs and flickr. Institutions could try to control what happens but there is a greater opportunity for them to both collect resources from their community (a list of academics on slideshare; an aggregated feed of departmental delicious or connotea tags) and make sure their resources are pushed out to other services too (publicising branded resources, offering training on web2.0 tools,  making sure institutional repository content is also push to relevant subject repositories).
9. Branding, publicity, goodwill
   • 7 and 8 create a lot of this; and I think as a community we’re beginning to understand this (for example: Library of Congress on Flickr, the various Opencourseware initiatives).I’m not saying it’s always going to make ‘commerical’ sense but sometimes it will and we’re getting better at remembering how much of our ‘income’ is public funding.
10. Services should talk to each other…
    • Many of the web tools and services that people use are beginning to offer the possibility of much greater connectivity. This ranges from connected, portable, or shared authentication/ identity (such as linking hotmail accounts; OpenId developments) to api-based access to content (such as: twitter/ ebay/ del.icio.us etc in netvibes, flickr photos pulled into animoto). I think there is a growing expectation that content and services should not be locked into one particular platform or interface.
11. Good metadata matters..
    • I’ve less direct evidence for this but I think that more we want to engage with all of the above, the more we need good metadata and more of it. The more tools and services scale the more that greater metadata quality is required (though see 1. this does not necessarily mean we should spend more time cataloguing). The clearest presentation I’ve seen of this in practice was in connection to IVIMEDS.