Sunday, 21 February 2016

Tim Worstall (and nearly everyone else) is wrong about open access publishing


I get it that Tim Worstall as an independent researcher is on the sore end of the historically dominant publishing business model. To remind you, this model is where the authors of research get published for nothing, get paid nothing and the publisher flogs the journals back to academic libraries for expensive subscriptions. Hence this what Tim is celebrating:

Knowledge is a public good, such research papers are meant to be read to spread it and almost all of the research was tax funded to boot. It does seem odd there’s a there’s a few gatekeepers waxing fat off the journals.

Now I'm not going to get into the economics of all this so won't be asking whether or not knowledge is a public good. However, it is moot whether what we buy from academic publishers is knowledge per se or whether what we're buying is the publishing process. After all there's absolutely nothing to stop researchers doing what Tim does and just plonk their work on a free to air blog. There won't be any peer review, it'll be hard to find and it is tricky to protect against copying or plagiarism but authors can do this right now without restriction.

Unless they choose to submit is to a recognised journal so as to secure the peer review, the editing, the abstracting, the writing of key words, the indexing, the dissemination and the guarantee that the work will be seen alongside other similar work with the assurance of quality. Plus - since the copyright has transferred to the publisher - the assurance that copying and plagiarism will meet with a robust response. Under the terms (they vary from publisher to publisher, these are Elsevier's) the author will be able to:

Share their article for Personal Use, Internal Institutional Use and Scholarly Sharing purposes, with a DOI link to the version of record on ScienceDirect (and with the Creative Commons CC-BY-NC- ND license for author manuscript versions)

This means the author can but a copy onto a personal site, can use the article in teaching, can share the work with other researchers and can put the article into an institutional repository. All of which uses are open and sit outside the article cost or journal subscription. Nearly everyone who wants or needs to view the research can do so. Indeed, for credentialed science journalists the big publishers mostly give free access to published research and nearly all research is free to air within six months to a year following initial publication (in Elsevier's case this extends to post-docs not in academia).

So wholesale theft aside, what's the alternative to this model? This is what's called open access and, far from being free, it replaces subscriber pays with author pays. And, because the authors of articles in academic journals are academics, they aren't about to start paying personally for something they've had for free up till now. So the 'author pays' is, in reality, 'author's institution pays' or 'author's funder pays'. All we have done is to shift the cost from the library budget to the faculty budget and research budget. And the cost is anything up to £3,000 per article (depending on the particular journal) so it's not really accessible to the independent researcher in the way the old system was accessible.

The bigger problems with open access publishing, however, relate to who gets published and the quality of the published work. Firstly, the control of publishing shifts from the editorial boards and academic peers to either the institution or the funder (most typically a government body or a private corporation). An academic can only get published what his institution is prepared to pay for since they have to stump up the three grand to pay the publisher for the privilege. We move from the system where the gatekeeper is the publisher to one where, overwhelmingly, it is the government or an institution dependent on the government for its funding, authority and existence.

The second issue is one of quality - here's Beale's List:

This is a list of questionable, scholarly open-access publishers. We recommend that scholars read the available reviews, assessments and descriptions provided here, and then decide for themselves whether they want to submit articles, serve as editors or on editorial boards.
The worst of these publishers are indistinguishable from a vanity publisher - except that they're exploiting institutional budgets rather than a personal, individual wish to be published. Other simply play fast and loose with the peer review process, with systems for citation and loopholes within the infrastructure around open access publishing. Also, given the way in which people (and especially non-academic people) search, there is a big risk that poor quality, under-reviewed research is given the same credence as high quality, fully reviewed research.

Here from one publisher is an example:

You may suggest information of some particularly qualified reviewers who have had experience in the subject of the submitted manuscript, but who are not affiliated with the same institutes as the contributor.


You may also submit a list of reviewers to be excluded.

There's nothing wrong with open access systems but we need to understand that this isn't about whether knowledge is a public good but rather about whether the value added by the publishing process - however it's paid for - is worth the price. The simple 'research should be free to everyone' argument is essentially identical to the 'free access to medicine' argument - it may be right, even a good idea but it certainly isn't free. Which is why Tim Worstall - and many others banging on about open access and 'free knowledge' - are wrong.



Kevin Kirk said...

I enjoy your blog and usually agree with you on most things. But on this I'm not convinced you are right. My job, at my university, is to review and edit journal submissions (mainly medical research) for international journals. As I see it the problem with having the research hidden behind expensive paywalls is that the researchers and/or journalists can make wild claims, often at odds with the findings, that very few people can challenge, unless their organization pays to get access. How many times have we seen 'research has shown' in the newspapers where the reality is that the actual research shows very weak correlation and certainly not causation; but, it fits the funder's views or intent (such as governments or single issue pressure groups) so it is used to justify certain governmental measures (alcohol pricing and e-cigs are two such examples).

The peer review system generally consists of you review mine and I'll review yours so it results in a lot of nonsense (mainly in 'public health') being labeled 'peer reviewed' even it is junk (too few research subjects, poor control, correlation being confused with causality and generally poor research methodologies).

Some universities are now going over to creating their own journals, which may be a way out of this impasse. These are mostly open access and are paid for out of university funds as part of the system - to be honest, creating a journal is not expensive. The system also benefits from the integrity of the university rests on the quality of both the published research and the peer review so it is in their interests to be rigorous as it would encourage serious researchers towards serious universities.

Chester Draws said...

Wrong in most regards.

Why should a paper cost £3,000? You're citing an old style publishers "Open Access". That's not even close to actual open access. Wiley are fighting the changes, not helping them.

The big academic publishers will publish anything that brings them money. Elsevier publish "The Journal of the Faculty of Homeopathy". It's complete rot from cover to cover, and certainly not science. But hey, it brings in a buck!

For every "quality" journal published, they publish much many more that are poor quality.

there is a big risk that poor quality, under-reviewed research is given the same credence as high quality, fully reviewed research

"Review" doesn't mean very much. It means that it meets minimum standards of layout etc. Too much stuff that is wrong, even fraudulent, gets past "review" to give it credence. Meanwhile often quite good stuff doesn't pass review because it rubs people the wrong way. In any field I have ever had any deep interest in I have learned that "review" means very little with respect to quality.

It's increasingly clear that most research published is wrong. Defending a system where being wrong in the "right" journal is better than being right in the wrong journal, as we have at present, is defending a system that is corrupt.

Anonymous said...

I do have to agree with the comments above. Science and technology, as reported by the press, is frequently written up by people with, at best, a superficial knowledge of the topic. Its primary aim seems to be eye-catching headlines. As things stand at the moment, it can be very difficult to access the original research and find out for oneself. Then again, perhaps I'm one of those weird people that likes some background to the headlines!