David A. Wheeler's Blog

Sun, 13 Dec 2009

U.S. research should be open access

The Office of Science and Technology Policy (OSTP) has launched a “public consultation on Public Access Policy”, to see if research funded by U.S. grants should be made available as open access results. I think this is important — I believe publicly-funded unclassified research should actually be made available to the public.

Historically, the U.S. pays a fortune for research, the results are written up as papers for journals, and then various publishers acquire total rights to these papers and charge exorbitant monopoly fees for them. The result: Most U.S. citizens cannot afford to see the research their taxes pay for.

The basic question here is really straightforward: Should publicly-funded research results be made available directly to the public instead? Or, should private companies continue to gain ownership over publicly-funded results, for either nothing or a tiny fraction of the public’s costs?

A small number of journal publishers and societies strongly want to keep things the way they are, of course. It makes sense from their point of view; everybody likes free (or nearly free) money! Historically, this arrangement was created because it can be expensive to publish and manage paper. However, that rationale has become completely obsolete. Few people want the paper any more — they want the research, on-line, without a paywall. And don’t give me nonsense about the “costs” of peer review. Many journals don’t pay their reviewers (the reviewers do it gratis), and even if they did, the total control they gain is still unjustified; the U.S. government spends far more per paper than they do for review.

The current sequestering of research is not good for science or the country. I’m currently reading the interesting book “Are We Rome?” by Cullen Murphy, and I can’t help but see some parallels. Chapter 3 is all about “when public good meets private opportunity”. Private organizations may pay for private research, and then keep their results private. But when the public pays for research, it should be shocking if it does not get released back to the public. And by “released back”, I mean released back at no fee at all.

So who will pay for the printing, complex peer review, storage, and fancy indexing of these research results? I think the very question shows a failure to understand current technology, but let’s answer it anyway. Most peer review isn’t paid-for anyway, and if it is, it’s a tiny cost compared to the research itself. Storage? Don’t make me laugh; for $100 I can buy storage for the all of the U.S. research papers for a year. Indexing? The government shouldn’t be doing serious indexing at all!! Just put it on a government site with a basic form filled out (title, authors, date, keywords, abstract, and a link to the actual paper on the government site). If it’s not behind a paywall, the many commercial search systems will index it for you.

I do think there should be a centralized government repository of such papers. If it’s distributed, then papers could be lost without anyone knowing it. I think they should be freely redistributable, so others can copy what they want, but a centralized repository would make sure that we keep all of them available forever. Also, bandwidth costs can be reduced by scale. There’s a risk that they all get lost at once, but it’s easier to copy everything if there’s one place to start from. If it’s a complicated site, then they’ve done it wrong…. for each paper there should be a simple “summary” page with title, authors, etc., and the actual paper itself.

OSTP cites the experience of NIH; NIH did wonderful work for releasing as open access, and in my mind the real problems are that they didn’t go far enough. First, NIH has a one-year embargo… if I already paid for it (and I did), why should wealthy people and organizations get the results first? Second, NIH only considers the actual papers, not the data and software programs that support the works… yet often those are more important. If they were funded by the public, then the public should get them (unless they’re classified, of course, but then they shouldn’t be released at all). I’m sure there are complications and exceptions, but a “default open access” policy would go a long way.

So please, tell the OSTP that the U.S. should release government-funded research as open access publications, available to anyone on the Internet without a paywall. In short, if “we the people” paid for it, then “we the people” should get it. For more information, see this Request for Information (RFI)

path: /oss | Current Weblog | permanent link to this entry