petermr's blog

A Scientist and the Web

 

Open Scholarship means Better Science

Four years ago [1] Open Access publishing was described by some members of the publishing community as “junk science”, the implication being that Open Access led inexorably to lower standards of (or even no) peer-review. I now assert that, from my perspective, Open Scholarship means better science, and invites readers to confirm or challenge this view. I use Open Scholarship to mean at least OABCD = Open Access, Bibliography, Citations and Data).

I now generally only publish in open Access journals – by this I mean “gold” Open Access where the author or funder pays for each submission and all articles in the journal are open. IOW I can go to any BMC, or PLoS journal and know that all the articles can be downloaded , read and re-used without further permission. Moreover I can build robots that systematically read every article in a journal. For example our robots have downloaded and understood 10,000 (sic) articles from the International Union of Crystallography’s Acta Crystallographic E.

The systematic retrieval and analysis of articles is critical to modern science. It is only possible with “gold”. It is forbidden by contract to use machines to read subscriptions to many major publishers, who – dog-in-the-manger-like – stop us innovating but do none themselves – these publishers do a huge disservice to science for the benefit of their shareholders and CEO’s incomes. The hybrid journals (where some articles are Open Access) and useless for systematic study as it is impossible to know which articles can be used for which purpose. The use of “green” publishing (where authors self-archive in repositories or web pages) is irrelevant as it is impossible to discover these publications systematically. [For example it is impossible to answer the question "find me all green-published articles in synthetic chemistry"].

I shall return to this in future posts – in this one I outline the virtues of Open Access publishing of single articles. I shall first put the counter-arguments (that closed access publishing is superior to open Access). They are the *only* arguments in favour of Closed access publishing:

  • Closed access publications have a higher standard of peer-review and editing. It is the only argument that could lead to the denigration of Open Access – that more bad science was allowed through the review system. I know of no objective evidence for this, other than proof by assertion.
  • Closed access publications have higher impact factors. This is difficult to measure. There may be some historical hysteresis in the system. Since, as I understand it, there is no objective measure of the impact factor (Bjorn Brembs outlined at OKCon2011 that how IF’s were calculated was a matter for private negotiation for each journal). Certainly the decimal points on IFs are ludicrous. IFs are, in any case, one of the worst measures of value. But the assertion may, for the moment, be true. Does this lead to better science? It is difficult to see how. It could conceivably lead to better targeted funding (“let’s only fund science reported in high-prestige journals”, but I know of no funders that take this approach – and rightly so.
  • Only closed access publishers can sustain the economy of science publishing. There seems little likelihood of Open Access publishers suddenly crashing from the marketplace. The new Wellcome/MaxPlanck/HHMI publication(s) will ensure very high-quality Open Access publishing.

So there is no clear benefit to *science* in choosing closed access. (There may be a benefit to individual scientists). I imagine that Closed access publishing has higher costs than Open access as it has to employ police to detect and chase people “stealing content” – I have no idea what percentage this is. Otherwise costs should be independent or closed/open and depend only on efficiency and whether the organization is for profit. These are difficult to argue.

So now the advantages to Open Access. I have submitted 11 papers or mine and 4 others to BMC, a “gold” publisher. Here are my immediate benefits:

  1. The manuscript gains earlier priority. I can claim 2011-07-05 as the deposition date for Open Bibliography in STM (http://www.dspace.cam.ac.uk/handle/1810/238406 ). If no-one else has published an article called “Open Bibliography” then we can claim priority.
  2. The manuscripts gets feedback. We have already got feedback on one, where the author of one of the programs we have used has pointed out that we have used an early version of the program. As a result we have re-run the analysis with the latest version and improved the result. This would not normally come out of most closed reviewing.
  3. We advertise the work we have done, even before the papers are published.

There is nothing absolutely novel in this – it’s what happens in arXiv – but there are publishers in chemistry which would immediately reject the manuscript unread if there was an already exposed version on the web. There is no *scientific* reason for this, only that it protects the closed access publishers’ business.

  • When the articles are published many more people *can* read them than if they were in closed access journals. This is incontrovertible – “all” is greater than “some”. The closed access argument is, I assume, that because closed access journals have (in their eyes) greater prestige, then more people will read them. I remain to be convinced, and would need firm evidence.
  • People can re-use my material without permission. The most valuable re-use is probably analysis and indexing by robots.
  • People can detect whether my work is valid. The more people read an article, the more likely errors are to be detected. Closed access has many fewer people who can detect errors.

 

If the academic world, in its inward-looking and self-congratulatory manner, continues to build self-perpetuating reward systems by promoting “high impact” brands, there will be an increasing clash between the need to develop science Openly and closed publication. But the branding system by itself does nothing to promote better science.

So the equation is simple. IF there is no difference in the quality of service provided by closed and open access science, then there is no intrinsic differential benefit from the publishing process. Open access is then better because more people can read the science and there can be much more re-use.

It is up to the closed access publishers to make an objective case why they provide a better service. They are welcome to post that case here.

 

 

[1] Four years ago a group of publishers, through their association AAP, launched PRISM – an effort to position closed access publishing as high quality and open access as leading to a number of evils , communally referred to as “junk science”. They hired a consultant who was well known for discrediting people and organizations (http://www.nature.com/nature/journal/v445/n7126/full/445347a.html ).

The consultant advised them to focus on simple messages, such as “Public access equals government censorship”. He hinted that the publishers should attempt to equate traditional publishing models with peer review, and “paint a picture of what the world would look like without peer-reviewed articles”.

And

Dezenhall [the consultant] noted that if the other side is on the defensive, it doesn’t matter if they can discredit your statements, [Susan Spilka, Wiley's director of corporate communications] added: “Media messaging is not the same as intellectual debate.

PRISM seems no longer an issue (its last news item was in 2007 http://www.prismcoalition.org/ ).

 

Leave a Reply