Universities as publishers?

UCL has become the first large UK HE institution to mandate Open Access. Here’s part of Nature’s account (reproduced with thanks but without explicit permission under fair use):

University College London (UCL) has become the latest institution to adopt an open-access publishing policy, adding to a rapid increase in such mandates over the past year.

Open-access analysts say the move foreshadows a series of announcements from many other UK universities that have been considering similar policies.

Under UCL’s system, all research published by university staff will be placed online in an institutional free-to-access repository — but only when publishers’ copyright rules allow.

UCL, one of the UK’s leading research-intensive universities, announced on 3 June that it had established a publications board to implement the policy. The policy will take full effect with the beginning of the 2009–2010 academic year, says Paul Ayris, director of UCL library services.

An important comment is added later:

UCL’s move is unlikely to improve public access to scientific research papers, as national bodies that support research already demand that. Thirty-six of them — including the US National Institutes of Health, all seven UK research councils, and the European Research Council — require work they have financed to be made publicly available (usually through deposition in open-access repositories such as PubMed Central, six months after publication).

But Alma Swan, a consultant for Key Perspectives, which analyses scholarly communications, says the recent flurry of institutional activity has come because university officials are realizing the importance of increasing their institution’s visibility on the internet, and of creating a complete record that can be analysed and compared against other institutions’ outputs or easily entered in national funding competitions. The UK and Australia, which both allocate funding depending on the quality of published research, lead the world in open-access repository policies, Swan notes.

[and, in passing, my congratulations to Paul and Alma for their dedicated efforts in this arena.]

The key point is that Openness on the web is a critical factor in institutions advertising their output. Conventionally a scientist’s impact is measured by the integral of their papers multiplied by the impact factor of the journal in which the papers appeared. This depends on how often the papers in the journal as a whole were cited. This is clearly an extraordinarily blunt metric – some papers are never cited and some are cited 10,000 times (normally in a ritual of copy and paste). The average figure for “high-impact” journals is probably around 5 (i.e. 5 citations per paper). That figure only appears some years after the journal starts.

In practice many scientists now discover their reading through Google and click on titles that look interesting. That’s a “download” and most journals have “highly accessed” papers. If you can trust the publisher to report accurate figures – the key word is trust – then it’s a more accurate metric. Still not very good, as many readers will flick between papers – for example I clicked through 3 papers in 10 minutes this morning just looking for a particular bit of data. The ones that didn’t have what I wanted will count as accesses – motivation cannot be recorded easily by weblogs. (I am sure someone will tell me that – yes – the clicks are all monitored and they know what’s going on in my brain).

So discoverability through Google is the primary way that scientists access information and find research papers. Let’s think what those scientists actually want. Sometimes they know exactly – the melting point of that particular compound. How to express that protein. But often they want to read a collection of papers relevant to a particular topic.

And they will find that half the papers they want to read cannot be read because their University does not buy that journal (correction – does not pay an annual rental for the journal). And, if they are in the charitable sector – patient disease groups – they may not be able to read a significant amount of material at all.

Now if a particular University was the world leader in that area it would make sense to collect all their public work in one place where it was publicly visible. It would be the natural place for people to look first. If another University felt this was unfair they would have to try harder to advertise their work. And so on – and every step of cometition would increase the visibility of the science. To anyone in the world who was interested. Anyone who had access to the Internet, agreed, but without any toll-barriers.

In essence the Universities would become Open Access Disseminators. That’s where current Closed-access models fail – they restrict dissemination. There is still absolute need for peer-review – and it’s unlikely an University can easily peer-review its own work. (It would be like Mps punishing themselves). So the current scholarly publishers could – and should – concentrate on peer-review.

But doesn’t this sound like an opportunity for a renaissance of University Presses? The costs of publishing have dropped – there are many small journals which are run at near zero cost, although larger journals with high rejection rates require paid staff. If we were starting today, in an e-only world, would Universities re-invent for-profit publishing?

Possibly. But it would look very different from what we currently have. And UCL, with many others can show the way.