A febrile vision of the future

Unfortunately I am not around for day two of intraconf10 but this is a quick (and slightly tongue in cheek!) contribution for the Forward to the Future Discussion Panel that will close the conference on Thursday afternoon.

The panel:

Ian Watson, IRISS
Lorna Campbell, JISC-CETIS
Charles Duncan, Intrallect
Neil Livesey, Learning and Teaching Scotland

The brief:

During the final discussion panel we will discuss visions of the future. Rather than consider what you think might actually occur we would prefer you to envision what you think the future of sharing digital resources should be like.

Twitter tag:

#intraconf10

---

Well…

A few over-riding issues come to mind:

• IPR/licencing – pain in the proverbial; will the law be rewritten one precedent at a time?
• Revenue streams – business models will have to change
• Metadata – will we still need formal metadata at all?
• Cultural change/zeitgeist – it’s already changing and will continue to do so as the Google generation comes of age.

And perhaps to separate out from these, there are also interesting questions of how the infrastructure and indeed the web itself will continue to develop:

• The semantic web – will it ever be fully realised?
• Google – will it actually take over the world or be sued out of existence (unlikely)?
• The Cloud – likely to have an increasing impact on (institutional) IT infrastructure

I don’t know if it’s this blasted cold but, as I steam through the Scottish countryside on a Virgin train, I’m finding it difficult to take the mental leap implied by the brief to what *should* rather than what might actually happen. Maybe thinking about the rail infrastructure as a metaphor will help…

In the future the railways will be so cheap to build and maintain that people will expect to travel for next to nothing; they will be fast and efficient (mostly using super-conductors to glide on a cushion of air) and you will be able to get anywhere in the country within half an hour. In actual fact, “railway” is a bit of an anachronism and fairly soon you will be able to take a train to the moon (run by Richard Branson and Google, obviously). This will take a bit longer and there will be a variety of “value added services” to keep the punters entertained for the 12 hour round trip (which will be an important revenue stream to supplement the nominal moon-train fare).

Okay, may be that didn’t help after all but the point, if I have one, is that Universities, if they are to compete in the future, will simply have to share their digital resources for free (whatever precedents of IPR are breached along the way); they will need to create new revenue streams and develop new business models, perhaps based on value added services because their customers (the Google generation) will be used to accessing everything for next to nothing. They will still be using Google, (who still won’t be supporting OAI-PMH) and whose kidult engineers will have cracked the semantic web thereby rendering formal metadata irrelevant. Moreover, Universities will have to get used to storing their data in the cloud and handing over control of that data to a variety of third parties.

What part will repositories play in this brave new world? Well, they won’t need to support LOM or any other metadata standard (except may be user tagging for pictures and stuff) and instead, the emphasis will be on disaggregated instances of a super-portal that uses a secret, post Pagerank algorithm to return the best possible resources from the Cloud to an information illiterate user!

The Digital Economy Bill notwithstanding.

Advertisement

Four JISC repository infrastructure projects

I was contacted this week by Evidence Base at Birmingham City University who are conducting a “short lightweight review” of four key repository infrastructure projects, preliminary to a larger evaluation of the IE programme as a whole, and are talking to JISC programme managers and project managers as well as seeking views from lowly repository managers like me!

The four projects I was asked to discuss were:

Repository Search (UK Institutional Repository Search- IRS) – http://www.intute.ac.uk/irs/
Repository Support Project (RSP) – http://www.rsp.ac.uk
Repository Junction (Open Access – Repository Junction – OA -RJ) – http://edina.ac.uk/projects/oa-rj/ and
Repository Aggregation (RepUK) – http://www.ukoln.ac.uk/projects/repuk/

Now I like to think I’ve got my ear to the ground and I was immediately struck that I was only actually familiar with two of these projects (the intute IR search and, of course, the good old RSP).  So I followed the links for the other two projects to learn what I could – both of which, in my view, need to be very much more high profile than they are currently (though they do both have another 12 months to run until 31st March 2011.)  My ensuing discussion with the lady from Evidence Base was more around the conceptual value of all four projects.

OA-RJ

I expect that OA-RJ in particular will gain traction over the coming months, not least because it is referenced in the current JISC Grant Funding Call Deposit of research outputs and Exposing digital content for education and research.

The purpose of the project is to scope, build and test a deposit broker tool to assist open access deposit into, and interoperability between, existing repository services; currently multiple-authored journal articles are deposited singly in either an institutional, funder or subject-based repository and the primarily aim is to simplify the repository deposit workflow for multiple-authored journal articles; OA-RJ will therefore offer an API that supports redirect and deposit of research outputs into multiple repositories.

RepUK

I was particularly interested in RepUK and IRS as I have for some time been a little non-plussed by our collective, continued obsession with the woefully under-used OAI-PMH and both these projects are using the protocol (I think!).

There is not a huge amount of information on the RepUK website but the paragraph below gives a flavour of the project:

“The interest in exploiting the content to be found in institutional repositories is growing. At the same time, there is a range of possible uses for a central cache of metadata records held by institutional repositories. Most notably, with a recent emphasis on ‘rapid innovation’, there exists an opportunity to position this aggregation of data to support research and development generally in the fields of metadata and/or repositories. Rapid innovation projects which require a corpus of metadata to work with will benefit from this readily available data-store, avoiding the resource-intensive overhead of developing their own harvesting and aggregation solution.”

RepUK also invokes Lorcan Dempsey’s concept of ‘concentration’ in a Web 2.0 environment as a “major characteristic of our network experience” involving “major gravitational hubs” that “concentrate data, users (as providers and consumers), and communications and computational capacity” and posits that “a central cache of metadata records held by institutional repositories” in this way, exposed by a simple, RESTful API, would allow the community to start building value added services around this (hopefully) high quality metadata.

UK Institutional Repository Search (IRS)

This service has come to the end of its funded period as a JISC project but is being maintained at a basic level by Mimas. I presume that it is using OAI-PMH* to cross-search UK IRs and offers “conceptual search” and “text mining search”**. With the best will in the world, it is difficult to see how this facility can compete with the likes of Google in its current incarnation

* May be conceptual search used OAI-PMH but “text mining” is more Google style?
** At least it did but the text mining search was broken and was giving a “Bad Gateway” yesterday – it now appears to have been rerouted to the “conceptual search” only, presumably while it is fixed.

---

Google, of course, withdrew support for OAI-PMH back in April 2008 and though I’m aware of a few harvesters around like OAIster, even OpenDoar uses a Google custom search – http://www.opendoar.org/search.php, not OAI-PMH, to search repository content.

I can offer only anecdotal evidence but I’m pretty sure that your average academic will tend towards Google/Google Scholar to source research on the open web and has no idea about the OAI-PMH which simply isn’t widely used enough to justify our ongoing fixation.  The  reasons for this are severalfold and represent, to some extent, the protocol’s pedegree (that dates back to the earliest days of the open access and institutional repository movements) and the associated investment by the community, in software specification for example; also from a recognition of the limitations of Google for academic purposes and the undoubted potential of OAI-PMH (though this potential has arguably been watered down by so many repositories also carrying metadata only records rather than exclusively full text.)

RSP

When I was new to repositories I found the RSP absolutely invaluable as a source of information and support, they came to a soft launch of our repository back in 2008 which was really useful to give colleagues a little bit of a wider view of the repository landscape in the UK. I must confess that I haven’t been back to the RSP website for a while and I was pleasantly surprised that there is now a great deal more content covering everything from a primer on the OAI-PMH to advice and resources for successful advocacy.  I was also reminded that the RSP do outreach visits and I may well consider giving them a call – it would certainly be useful to get an objective perspective on some of the issues we continue to face with repository development here at Leeds Met.

I’m not naive, of course, to the reasons for JISC conducting these project evaluations and they clearly want to think carefully about where future investment can most effectively be made; I was asked a few leading questions around how the RSP still meets the needs of the community (I think they do!) and how they might adapt their approach to meet shifting requirements – the start-ups are all but finished I think but, no doubt, new people are coming into the sector all the time who will most certainly benefit from the clear information and support of the RSP.  I also speculated somewhat idly whether the website could be a bit more dynamic and, well, Web 2.0 – they do have a presence on Twitter – @RepoSupport – but I couldn’t find it from the website and I don’t think it feeds there.  I even wondered whether a social network style site using ning or elgg might work….just a thought.

What would you like to search for – research or OER?

Leeds Met Open Search – http://repository.leedsmet.ac.uk/main/index.php – now incorporates a “splash screen” that allows the user to choose which collection they wish to search with links that provide access to separate interfaces that are tailored to each type of material:

Each tabbed interface provides an appropriate Advanced search form as well as relevant browse options; by LCSH or faculty for research and by HEA Subject Centres or JACS code for OERs:

Once again, massive thanks to Mike for his rapid response to the the myriad requests I make of him on a daily basis!