Sunday, March 15, 2009

Open wider

There seems to be a bit of a discussion between Joe Esposito and Stevan Harnad on Liblicense, loosely about the significance of OA and of peer-review.

Quoting Joe Esposito (reacting to an article by Richard Poynder on 'Open and Shut'):
"The real thrust of the world of open access is neither green nor gold, but what I have termed "unwashed," that is, the vast and growing – and growing and growing and growing – world of material that is not peer-reviewed. Take Poynder's own article, for example, or posts to this list. Look at the material that is accumulating in IRs, arXiv, and elsewhere; think about all the blogs and Twitter feeds.

The evidence is mounting that many advocates of open access have never actually used the Internet. The myth persists that OA publishing is just like traditional publishing except that it is free to the user. While there are some segments of OA that are just that, it is a shrinking part of the open access material that is being generated. And it is minuscule compared to what we will see in the years to come.

This doesn't mean peer review is going away. It simply means that peer review is evolving to conform to the characteristics of the online medium, just as the novel grew with the printed page and tennis is a game played around a net. Increasingly peer review will be post-publication, not pre-publication. I suspect all this talk about Gold and Green is a waste of everybody's time."
Quoting Stevan Harnad (reacting to Joe Esposito):
"Or could it be that some of the opponents of Open Access to the 2.5 million articles published annually in the planet's 25,000 peer-reviewed scholarly and scientific journals have never actually done any scholarly or scientific research, hence never published in a refereed journal, and never had any need to consult one for their scholarly and scientific research?"
Open Access to peer-reviewed material is important, but to reduce the scholarly knowledge exchange to just peer-reviewed articles is to ignore the massive amounts of data and knowledge that are shared in other ways. I see the importance of unrefereed scientific published material increase. Dramatically. As long as it is open.

But what does that do to the trustworthiness of that information? Isn't the whole point of peer-review to make sure that what is published conforms to accepted standards of scientific inquiry so that the reader can have a certain amount of trust in the results that are presented? Well, of course. But what is presented in journal articles are mostly results derived from data. Interpretations and annotations of data. Seldom the data themselves. Journal publishing evolved in the past, when the physical reality of sharing actual raw data was nigh impossible, so almost every scientist had to rely on the interpretations as published in journals. But now that we can share the raw data (view Tim Berners-Lee's call for sharing raw data), and tools to manipulate those raw data become widely available, relying on journal articles may well take second seat. And now that instant comment on data as well as on journal articles has become possible, with blogs, twitter, and what not, review after publication is a reality of today (albeit not used all that widely yet).

Furthermore, technology is emerging that is able to quickly identify if data and articles are in essence in line with the scientifically accepted knowledge of today, and is merely confirmatory in nature, which makes the outliers stand out. Those can either be scientific rubbish, or potential breakthroughs, and a peer-review process is well-spent on them, ante (the technology is a great tool for editors!) or post publication.

Peer-review may or may not survive in the way it is now. But it seems clear to me that openness of published articles as well as raw data is, after initial hesitant steps, bound to show explosive growth.

Jan Velterop

Monday, March 09, 2009

Sunday, March 08, 2009

Getting the right arguments right

Congressman John Conyers has publicly responded, on the Huffington Post, to the call from Larry Lessig (initiator of Creative Commons) and Mike Eisen (initiator of PLoS) to speak up.

Peter Suber, in turn, responded in detail to John Conyers. Admirable detail, and a discussion with well-articulated arguments, like this, is the way forward, in my view. In an earlier post, 'Aiming at the right target', Peter is saying "Let's not make it easy for the bill's supporters to say that the critics simply don't understand". He is right, and in that vein I feel that I should humbly offer some advice.

In one of his arguments he points out the problem with the old NIH policy, which, he says, "...had the effect of steering publicly-funded research into journals accessible only to subscribers, and whose subscription prices have been rising faster than inflation for three decades". It is the second half of this sentence that is misleading. Technically it is true, of course, especially if he refers to average prices. But it is misleading by omission.

There are at least two reasons why the comment about inflation has to be put in context in order to avoid being misleading:
  1. The average journal prices have risen faster than inflation, but the average number of articles published in them as well, reflecting the above-inflation rise in scientific output. The correct measure should not be the average journal price, but the average price per article published. That may still have risen faster than inflation, and I haven't done the math, but having those data would turn the argument of inflated prices into a real one, or render it irrelevant.
  2. Secondly, scientific journal publishing is a global pursuit. Inflated prices may just as easily be an effect of a precipitiously plunging currency (at the library's side), or a steeply rising one (in the publisher's country), as of publishers' pricing policies. Indeed, if much of the work hadn't been sweat-shop-ized, outsourced to low-wage countries, price rises might have been much bigger. As with so many of the goods we purchase these days.
I think the arguments for open access are strong enough without the inflation red herring.

Jan Velterop

Tuesday, March 03, 2009

Footing the Bill

Should you need any further evidence that the American democracy is in essence a lobbyocracy, the anti-open-access bill of congressman John Conyers provides it. Of course it isn’t called the ‘Anti Open Access Bill’, but the “Fair Copyright in Research Works Act”. But then, this is the way of the world these days: euphemania.

The Americans are not alone in living in a lobbyocracy, where powerful special interests rule the roost. In other countries the people do as well. Take Australia. But Australians don’t seem to do euphemisms. They call a spade a shovel, and they have a web site to address these matters, unambiguously called, exposing money flows in politics. (By the way, and are still available today, March 3rd 2009, should anyone want to do the same in the US.)

It is amazing how misguided the reasoning is of Conyers' bill (Peter Suber does a sterling job exposing the fallacies in his Newsletter). "Fair copyright in research works", huh? For a scientist, fair copyright is a notion used to ensure attributed plagiarism, otherwise known as ‘citation’. It is one of the most important things about copyright. No, it is the most important thing about copyright. For a researcher.

For publishers it’s different. For them, copyright, or rather, the transfer of copyright, is a way of payment for the services they render. Though they call themselves publishers, these services are hardly to be called publishing any longer (in the sense of ‘making public’). They are procedural services resulting in the labelling of an article as ‘peer-reviewed and accepted by’ a given journal. The act of publishing is on the web these days, and anyone can do it. This is, of course, precisely the problem. The publishers’ business models are based on the idea that it is they who are publishing. They did, but that’s the past, when print was the only means of dissemination, of making public.

That said, the 'publishers' do fulfill a role that is needed in science. Researchers are required to publish in peer-reviewed journals. Essential for survival in the ego-system. ‘Publish or Perish’, remember? Of course, they also need to read, although the imperative isn’t quite there. No such thing as ‘Read or Rot’, after all. But to publish is the key to any career as a scientist at all. This fact should inform the business models: he who has the most interest pays.

Back to copyright. For publishers who think they publish, the transfer of copyright is just a way in which the author pays for the publishers’ services. If the value of that copyright is eroded – or in the view of some publishers even nullified – by funders’ mandates and embargoes, they have a problem. The most straightforward way out of that is of course substituting a monetary charge for the transfer of copyright. This is what the open access publishers have understood. The so-called ‘gold’ open access model.

So what should traditional publishers do? (I’m assuming that an egregious bill such as Conyers’ will fail.) Should they refuse articles that come with open access mandates attached? After all, they do not come with the required ‘payment’ of full copyright transfer. And embargoes are problematic (although the argument that articles have appreciable economic value after the typical embargo period of 12 months is rather weak, to say the least, seeing that almost all of the revenues of a publisher are realised in advance, as the subscription and licensing model demands). Refusing is hardly possible if they want to stay in business at all, since the authors are obliged by their funders to withhold transfer of copyright for anything other than temporary (a period of generally a year) and have no choice. Here, too, open access publishers have the advantage. After all, they simply do refuse articles that come without payment. With some discretionary exceptions, their policy could be expressed with the slogan “Pay, or just go away!”

But one thing I rarely see or hear. That is the notion that mandates with embargoes are a threat to ‘gold’ open access publishers as well. Especially the mandates with short embargoes, of, say, six months. What if researchers can wait that long to see most articles? And authors to publish their articles? Neither on the side of the reader or the writer would there be an incentive to pay for the necessary service that publishers do provide, be it in the form of transfer of copyright or plain money.

Which brings me to my final point. Payment for ‘gold’ open access publishing the way it is done now is also problematic. The reason is that payment for the services of a publisher is fully loaded on the published articles (and the same is true for ‘toll-access’ publishing as well, of course). And yet, much of the work is related to articles that do not come through the peer-review process and are rejected. A truly fair system would charge a submission fee, for which the publisher would organise the peer-review process. Like a driver’s test. You don’t just pay when you’ve passed and get your driver’s licence. You pay every time you take the test. It would probably also mean alleviation of the peer-review burden, since submissions would be carefully pitched to the journal of the appropriate level for the article, and not be allowed to cascade down the journal hierarchy.

Could that be a bill to put before Congress? Requiring that all scientific research is published with open access and that the only charges scientific journals can make are submission charges?

Jan Velterop