Wednesday, December 14, 2011

PDF resurrected

This blog is devoted to open access. Please subscribe. To the concept of open access to scientific information, that is.

Not for the sake of open access in itself. No l'art pour l'art. But for the sake of enabling scientists to make use of any information that is relevant to their research in any way that makes sense for them. In that spirit, please allow me to divert to writing about something that is not open access as such, but does help scientists to get to and use available knowledge more efficiently and conveniently.

Much – actually, the overwhelming majority – of the scientific literature is made available in the form of PDFs. There are good reasons for that. Easily downloaded, easily stored on your hard disk, easily printed nicely, integrity guaranteed to a satisfactory degree ('version of record'), et cetera. But in a web-connected world, having static, 'dead' documents like PDFs also has major drawbacks. Many scientists would like to look 'beyond the PDF'. Me, too. I am on record to have used the awkward verb 'depedefy' and making the case that that is just what should be done. No longer.

What has changed? Well, Utopia Documents. It is a scientific PDF-reader that connects the articles you have in PDF format to the web. Any PDF document that's not just a bitmap (image). The published articles, manuscripts written in MS Word that you have saved on your laptop or deposited in repositories as PDFs, even whole books. When you have Utopia Documents installed (it's free, and available for Mac, Windows and Linux from, the advantages of PDFs remain, and the disadvantages begin to melt away.

The current version of Utopia Documents is optimised for the life sciences – molecular biology, biochemistry, preclinical medicine, and the like. But this is clearly only the beginning. New functionalities and links to resources are continually being added and upgrades will be released regularly. Progress keeps being made in the foreseeable future and beyond, but that's no real reason to wait with using the PDF-reader, of course.

Conflict of interest: no conflict, actually, just interest. I'm a fan. I will do what I can to advocate Utopia Documents because I think it is a wonderful tool for scientists, potentially making their lives easier and their research more effective. And I and my colleagues will assist those who developed it and are continually improving it, with ensuring its sustainability.

Please help. The only things you have to do is to download the software, start using it, and tell your friends and colleagues about Utopia Documents.

Jan Velterop

Tuesday, December 13, 2011

Science Publishing: All About Submission

I think 'gold' open access publishing needs to be supported by submission fees rather than article publication fees, as is now generally the case.

The basic reason I am in favour of submission fees is that it makes scientific publishing really the service industry that it is, its main task nowadays having nothing to do with publishing per se, but mainly with arranging peer review and quality assurance of one sort or another. 'Publisher' is therefore a bit of a misnomer by now, a relic of the past. Publishing, as in 'making public', is very easy and people can do it by themselves, in a way that one does on a blog, for instance. Or even by depositing a manuscript in an institutional repository, which is publishing, in the sense of 'making public'. Science publishers should really be called 'quality assurance providers' or something in that vein. Because that is what a modern STM publisher is. A (perhaps too simplistic, but quite useful) model may be the 'exam' model. Submitting a paper is not unlike applying for, say, a driver's test, for which you pay, irrespective of the outcome. An article's scientific robustness is being tested; little else is of relevance (or rather, should be of relevance).

Apart from this, there are some clearly beneficial consequences of a submission-fee system.
  • It discourages spurious submissions and encourages pitching at the right journal at the right level
  • It therefore relieves pressure on the peer review system (fewer unnecessary rounds of peer review)
  • It relates any fees paid for the main work done by 'publishers'
  • It allows any prestige journals (insofar that they have a reason to exist) not to have to worry about high rejection rates and the related necessity of high article publication fees otherwise needed for OA to the small number of accepted articles (the Nature and Science argument)
  • It spreads the amount needed by a 'publisher' over a larger number of articles – the accepted plus the rejected – leading to the possibility of lower average fees
  • It removes the suspicion that OA journals might be tempted to accept more than they should just because of the money that accepted articles bring
To be fair, there are also downsides.
  • The need to be able to justify rejections properly, particularly if challenged (after all, submitters have paid for an assessment)
  • The reality that other publishers offer free submission (although this argument may not cut too much ice, given that it was also used against author-side payment, which turned out not as deadly to the model as was thought by opponents of OA)
The last point is probably keeping publishers from going in the direction of submission fees. I do hope that one of the more visionary publishers dares to make the plunge.

Jan Velterop

Sunday, December 11, 2011

Hybrid journals – double or quits?

Hybrid journals – journals that combine toll access to some articles with open access to others – do not generally enjoy a good press. Terms such as 'double-dipping' are used. This is not justified, as a general rule. I can't guarantee that double-dipping never happens, but I don't think it is generally the case. Publishers could do more to disabuse the library and research communities of the notion that it is, though.

That said, it is difficult, because even the basic understanding of how a subscription system works is often lacking outside (and even sometimes inside echelons of) the publishing community. One of the difficulties is that deciding on the price of subscriptions depends on a number of prior assumptions. There are possibly more than these three, but they are the most important ones: 1) how many subscriptions will we be able to sell; 2) how many submissions will we get and how many of those will be accepted for publication (i.e. what will the costs be); and 3) what margins can we expect to contribute to overheads and profit (or surplus, in the case of a not-for-profit publisher).

Typically, a publisher will have a portfolio of journals of which some do well, some just break even, and some make a loss if all costs, including overheads, are fully allocated. Hybrid journals will be found in all three categories. So what does 'double-dipping' mean? Are loss-making journals 'half-dipping'? Is 'double-half-dipping', in the case of those loss making journals, just 'single dipping'? Does it even make sense to think in those terms?

I think not. The objective way to look at it is to see the subscription price as the price to be paid for the non-OA articles that are published in a hybrid journal. That may be low or high if expressed in subscription price per non-OA article, but that is what a subscription to a hybrid journal is. Incidentally, comparing subscription prices per article (p/a) across a library collection will show a very wide range, and the inclusion or exclusion of hybrid journals is not likely to make any difference in the distribution of p/a in that range.

It may be helpful to think of a hybrid journal as twin journals sharing the same title, Editor, Editorial Board and editorial policy: one subscription-based, and one OA.

The OA articles in a hybrid journal are just as much OA as in any OA journal as long as they give the reader/user the same rights (of access and re-use), i.e. as long as they are covered by a licence such as the Creative Commons Attribution License (CC-BY) or the CC Attribution and Share Alike License (CC-BY-SA) and not the CC Attribution Non-Commercial License (CC-BY-NC). Sticking to CC-BY-NC licences, which does happen, is a sign of insecurity on the part of a publisher or of a lack of understanding as to what the purpose of open access actually is. Though there may be a number of cases where the publisher has overcome that insecurity but just hasn't thought about changing the licences yet.

As said above, hybrid journals do not generally enjoy a good press, but I have heard positive comments about them as well in the scientific community. Those relate to the notion that the editorial policy (the acceptance/rejection policy) of hybrid journals is not influenced by the potential financial contribution coming from the authors. The 'open choice' is typically given as an option only after the article has passed peer review and is accepted. I don't think acceptance and rejection policies of any respectable OA journal are influenced by the prospect of authors paying, and I certainly don't know of any such practices at the OA publishers I am familiar with, but it is an extra assurance hybrid journals offer that that is indeed not the case.

Jan Velterop

Saturday, December 10, 2011

The future of today is not what it used to be

One of the threads on the Liblicense-L discussion forum, on 'the future of the subscription model', has been running for quite a while now. Without much consensus. A few exchanges on the desirability and practicability of an 'author-side-paid' open access model as an alternative, strayed into a system financed by submission fees rather than fees for publication, and I am inviting views on that idea. The exchange on the forum, in chronological order (unedited save for a few typos and the insertion of a few hyperlinks):
(7 December 2011) There was an earlier comment on this thread (which I lost, alas) to the effect that one way to build an author-pays service is with a fee for submission rather than for publication. This is a great idea, and in a world without ruinous competition (John D. Rockefeller's phrase), it would work beautifully, as it aligns the cost to authors with the actual cost of delivering the service. But what happens when your competitor offers a free Christmas promotion? Or if eLife takes 10 years to figure out a business model? In a competitive market, you can never be smarter than your stupidest competitor, and if that
competitor wants to give away the store, I can see your store loaded onto someone else's truck.

Joe Esposito
(7 December 2011) Joe, isn't this already happening? And isn't this why a system based on submission fees hasn't successfully emerged yet?

The competition (subscription-based journals) are offering free promotions (to authors) all the time. They have found people who pay them through the back door (librarians, paying for subscriptions, as
long as it lasts). "In a competitive market you can never be smarter than your stupidest competitor." The words are yours.

This discussion is called "Future of the Subscription Model". The fundamental issue here is that the subscription model is simply not suited to an environment where maximum distribution is possible without marginal cost, and what is being distributed is not consumable (in the sense that it disappears if you consume it). In the bible there is a story about 'loaves and fish'. Allegorical (I presume). But scientific information in the internet environment is like the biblical loaves and fish. Albeit not food for the body, but food for thought. Scientific thought.

Jan Velterop
(9 December 2011) Oh, gosh, Jan, where to begin? This is just plain wrong. There is nothing "back door" about having a librarian pay for something. And it would be a wonderful world if maximum distribution were possible without marginal cost, but in fact there are huge costs to that distribution, if by "distribution" you mean that you persuade people actually to read something. Moving bits around costs nothing, and presumably this is what you mean, but the bits on my hard drive are meaningless unless I engage with them.

We have here the old saw about non-rival goods. It does not apply to media. Media is not a product but something that must engage human attention. That's a scarce thing. There is no superabundance of information when you take into account that someone has to be thinking about the information.

But it really is unfortunate that you insist on making this a binary game. I don't know if I could possibly have been more lavish in my admiration for the author-pays model.  It sits side by side with the subscription model and other forms of traditional (that is, toll-access) publishing. Who has to choose? Over time, the different economics of these models will influence the nature of the content such that you will get different things from subscriptions than you do for author-pays. Is there anything wrong with that? Why would anyone accuse a radio of not being a television?

Joe Esposito

(9 December 2011) Joe, I fear we are talking cross-purposes. My frame of reference is primarily STM journals. In that frame of reference I just don't recognise your definition of 'distribution' as "actually persuading people to read something", unless 'something' means literally that. Sure, publishers try to stimulate downloads, since they help them making he case to librarians that they should renew their subscriptions and licences. But it's a numbers game, in which 'something' pretty much means 'anything', and the marginal cost of extra downloads is negligible.

Indeed, the bits on your hard drive are meaningless unless you engage with them. Researchers do engage with the information they have access to, and they would like to have even more to engage with, but all this engaging isn't necessarily in the form of reading nowadays. Perhaps it can be described as 'meta-reading', but it is more and more about extracting facts and assertions, collating them with those from a large number of publications, connecting and relating them, analysing them, and using the information gleaned as a basis for further thinking, experimentation, et cetera. Occasionally articles are still being read linearly, but even then, particularly if they are being read online (or nowadays also in PDF when the PDF is opened with the likes of the scientific reader Utopia - free from as a starting point for further navigation of information and knowledge.

Human attention is indeed a scarce thing. And that attention is less and less being attracted by journals per se, let alone their publishers. It's the connections between facts and information across a vast array of publications and databases that attract attention. Fragmentation of information in all manner of different journals some of which are accessible and some not is the scourge of many a scientist. The observation that most articles are being accessed only after having been found with general search tools such as Google are testimony to the fact that practically nobody relies anymore on the choices journals and their publishers make.

The role of a publisher nowadays is to provide the service to a scientist of having his or her contributions peer reviewed and subsequently added to the common pool (well, ocean) of knowledge and information in a standardised, accessible and attributable way. That role is not satisfactorily played with the encumbrances of the subscription model.

You are right that the subscription model may be suitable to content of a certain nature. Magazines – even scientific ones – and the like come to mind. Nothing wrong with that. For the mainstay of scientific communication, however, the model is not suitable any longer. Of course it will amble along for a time. Quite a long time, even. Inertia is a pretty strong force.

Jan Velterop

The discussion is likely to continue...

Friday, December 09, 2011

Mandatory Academic Freedom?

From the Internet Freedom Conference in The Hague8-9 December 2011 (or from following on the internet) Dimitar Poposki (Twitter name @) sent the following tweet: "Q: Can Open Access to taxpayers scientific research be considered as a mandatory academic freedom?"

The answer is 'no'. 
First of all, what is 'academic freedom'? From J. Peter Byrne, "Academic Freedom", 99 Yale Law Journal 251, 252-253 (1989):
The First Amendment protects academic freedom. This simple proposition stands explicit or implicit in numerous judicial opinions, often proclaimed in fervid rhetoric. Attempts to understand the scope and foundation of a constitutional guarantee of academic freedom, however, generally result in paradox or confusion. The cases, shorn of panegyrics, are inconclusive, the promise of rhetoric reproached by the ambiguous realities of academic life.
The problems are fundamental: There has been no adequate analysis of what academic freedom the Constitution protects or of why it protects it. Lacking definition or guiding principle, the doctrine floats in law, picking up decisions as a hull does barnacles.
Einstein defined it rather pithily as:
"The right to search for truth and to publish and teach what one holds to be true."
And added to it a duty:
"One must not conceal any part of what one has recognised to be true."
In that vein, I would think it easy to make the case that another duty is publishing with open access any research carried out with public money.

But 'mandatory freedom' is an oxymoron. That's why the answer is 'no'.


Wednesday, December 07, 2011

Of Loaves and Fishes

And when he had taken the five loaves and the two fishes, he looked up to heaven, and blessed, and brake the loaves, and gave them to his disciples to set before them; and the two fishes divided he among them all. And they did all eat, and were filled. And they took up twelve baskets full of the fragments, and of the fishes. And they that did eat of the loaves were about five thousand men. 
Recognise this story? Well, it is about scholarly publishing. Avant la lettre. You realised that of course. 'Loaves and fishes' stands for 'food for thought': scholarly, scientific articles. You can share the information without losing it, until everyone hungry for it has their fill, and when they have, and start discussing, you are likely to end up with more information than you dished out initially.

The story is already some 2000 years old! Shame it took us so long to understand its meaning.

Tuesday, December 06, 2011

Copyright or Controlright?

Copyright is funny business. Even in science. Even in Open Access. You would be forgiven for thinking that copyright is all about protecting economic interests. But you would be wrong (though sometimes it is about protecting economic interests, honest).

Let’s examine the case of Open Access. Formal Open Access publishing that is, along the model of author-side payment for the service of publishing. It works like this: a number of authors have written an article (single authorship is quite rare these days) and submitted it to a journal that offers Open Access should the article be accepted after a process of peer review, for the quid pro quo of an amount of money. That amount of money compensates the publisher for the fact that he won’t be able to sell the article anymore (by way of subscription to a journal, for instance) when it is published with Open Access. After all, the article is openly and freely available (that’s what Open Access means) so who would buy it? Sound reasoning.

By the way, paying for the service of publishing one’s article is nothing new. In the past an author paid by transferring copyright (which the publisher could then convert into money by selling subscriptions – in effect selling access rights); in the Open Access model the author pays with actual money. Much more straightforward.

So far, so good. But you would think that if an article is published with Open Access you could actually use it. Simply reading articles is an antediluvian notion, and it is the re-use that makes Open Access worth it. Many Open Access articles can be re-used, and that is usually indicated by a so-called CC-BY copyright licence (Creative Commons Attribution), that requires, where possible, attribution of the authors, which is fair enough in the scientific ego-system. Besides, the authors (or their funders) actually paid for Open Access, so attribute is the least one can do.

But, alas, not all Open Access articles carry a CC-BY licence. Too many have a so-called CC-NC licence, which stands for Creative Commons Non-Commercial. This means that even the slightest possibility that re-use might result in some commercial activity, even if just derived in the second degree, re-use is prohibited. Goodness knows why.

What happens here is really not in the realm of protecting commercial interests any longer. Copyright has morphed from a way to protect legitimate interests into a way of controlling (read ‘restricting’) usage. Of Open Access articles! 1984 revisited! I’m sure it happens without many publishers involved even realizing.

C’mon publishers, the CC-NC licence doesn’t give you a penny more in revenues and it frustrates the hell out of scientists. Please change it to CC-BY. Please.