Open access is not such a smart idea

Elsevier ban

Should readers pay to read academic journals, or should the authors have to pay to publish their articles? (Graphic taken from

The recent move by the  Wellcome trust to create a new Open Access journal has highlighted a long-standing problem with the publication of academic and scientific works. The Wellcome Trust’s proposed solution has been reviewed favourably in all the newspapers I have seen, but I think the extent of the problem has been overstated, and there are serious problems with the proposed solution.

Let me explain:

  • Publishing journals is an essential activity for science. Journals form the archive of our collective endeavours and they are precious.
  • Publishing journals costs money and somebody has to pay for it.This can be either (a) Readers, or  (b) Authors. Option (c) would be ‘somebody else’ but there are no candidates that I know of who would be willing to do this. There are difficulties with (a) and (b) choices.

The problems with the current system (a) have been well-highlighted, but summarising, the main objection is that since the work was generally paid for with public money. the public should have the right of free access to it. What possible objections could there be to such an obviously fair proposal? Well, Slightly to my surprise I have four!

  • My first objection is that the arrangement by which someone who wants to publish something pays to have it published is well established: it is called vanity publishing. Another name for people who pay to have their work published is advertisers. Neither vanity-publishers nor advertisers are in general associated with high quality output. With the ‘author-pays’ dynamic, it is the author who is the ‘customer’ and in business, the ‘customer is king’. There is a serious danger that journals will lower the bar to publishing for those that are able to pay. In contrast, with the current system, anyone can publish for free – if the quality of their article is high enough – and so it is in the interest of journals to simply pick the best papers.
  • My second objection concerns the cost of making a journal article ‘Open Access’: typically £1500. This is easy to find for well-funded researchers, but it is not so easy for poorer researchers. Imagine a PhD student who wants to publish their work but whose supervisor objects? Imagine a researcher in a poorer country for whom the idea of spending £1500 is a dream! I know retired scientists who publish excellent work but who could not afford to pay for it to be published. And consider this: would Einstein have published all his papers in 1905 or would he have only submitted what he could afford? Since academics are judged in large part on the number of papers published (not necessarily a good idea but a fact of life) this would inevitably favour wealthier students and Universities. Is it not better to have all the information published with a criteria based solely on the quality of the work rather than just what people can afford to publish?
  • My third objection is that the problem with access is overstated. All the abstracts of each paper are available for free (e.g. here), and if you write to the authors they will usually be happy to send a pre-print of the paper – an unformatted version containing all the information and data – for nothing. I admit this is slower than ‘clicking’ and downloading – but it is free. Remember journals can only copyright the presentation of (say) a table of data, not the data in the table.
  • My  final objection is that if the budgets for science are fixed, then this proposal would cut the amount of money spent on science. If  (say) £100,000 of research produces 1 paper, then paying £1500 for the publication involves spending  £1500 less on the project – a 1.5% cut. If the research project was more successful and produced two papers – the cut would be 3%!
I am not denying that there are problems with the current system of academic publications – principally its expense, which makes access expensive (typically $30 per article) : academic publishing is run by businesses not charities or not-for-profit foundations.
Speaking entirely personally, my main problem with academic journals is the declining ratio of the amount of papers I read to the amount of papers I write.  I am under intense personal pressure to publish papers and increasingly I find it hard to find the time to read as many papers as I would like. If my experience is anything close to typical, then that bodes ill for the quality of published work.

Tags: ,

23 Responses to “Open access is not such a smart idea”

  1. Jan Velterop Says:

    Re objection 1.:
    a) Journals that like to be regarded for their prestige would commit suicide by accepting low quality papers, even if paid for by the authors;
    b) The acceptance bar for a large number of subscription journals is already very low;
    c) *All* articles are ‘advertising’ the authors’ scholarly prowess and *all* publishing is a form of vanity publishing in the scholarly ego-system (clocking up impact factors and the like).

    Re objection 2.:
    a) $1500 looks like a large amount, but compared to the average cost per article to Academia of the current subscription system – in the order of at least 4 times as high – it is a bargain. It will also decrease, since there is more real competition in an OA author-side paid system than in the subscription model;
    b) An author-side payment OA model may well have the effect of less ‘salami-slicing’ and informal (non pre-publication peer-reviewed) publication of minor results;
    c) Funders of academic research and institutions have always financially supported the subscription system (readers not paying anything) and it only stands to reason that they will also take on the – lower – cost of author-side paid OA publishing;
    d) The impecunious author without funders may indeed have a harder time to get formally published, but informal publication is an increasingly realistic option and in any event the plight of the poor authors have to be weighed up against the plight of the poor readers, the latter being far larger in numbers.

    Re objection 3.:
    The idea that articles are just read one by one by individual researcher may hold in some disciplines, but in modern science the help of computers is needed to ingest the information in the large numbers of papers published. This can’t be done without forms of text- and data-mining, which makes asking authors for copies of the full text once one comes upon an interesting abstract an endearingly old-fashioned idea, which is completely unworkable by now.

    Re objection 4.:
    a) The idea that author-side payment for publishing cuts into available funds for research, while reader-side payments (or rather, library-side payments) for subscriptions don’t, seems to be based on the premise that money spent on subscriptions is a different kind of money which cannot be used in any way for research. That money largely comes from the same funders as the author-side payments would, one way or another (e.g. via overheads institutions charge on grants), and so equally “cuts into the amount of money spent on science”.
    b) If research isn’t published it is not added to the ‘minutes of science’ and generally regarded as not having taken place, so the cost of publishing are integral to the cost of research.

    • protonsforbreakfast Says:


      Forgive my delay in replying: I have been overwhelmed at work and home.

      I think you may be wrote more than I did, so I won’t respond to all of your points individually: you make them well. However consider just this one point.

      Libraries with printed journals: do you want them to continue to exist? Is so, then who will pay for these to be printed? Will the author pay and then a publisher will distribute the journals for free?

      I object to the assertion that publishers are ‘evil’ and exploiting academics and keeping knowledge secret. There are real costs to paid and someone has to pay them. To my mind, the sustainability of alternatives to the current predominant system has not been demonstrated.

      All the best


    • Jan Velterop Says:

      Michael, as to your question whether I want print journals to continue to exist I’m agnostic. I don’t believe having print-only is sensible any longer. And print journals cannot be distributed for free, obviously. For a modern journal, Open Access to the electronic version is the thing to have, and possibly a paid-for print version as an extra service if there is enough demand for such a service.

      I also object to the idea that publishers are ‘evil’. They just make the most of their position in the scholarly ‘ego-system’. The viability and financial sustainability of OA publishing has by now been clearly demonstrated by PLoS and BMC for instance. The high profit level sustainability not, but high profit levels are normally a signal of lack of meaningful competition or market naivity.

  2. Anon. Says:

    Re: objection 2.:
    There are lots of price points out there already (esp. if you look outside of the PLoS/BMCverse). I currently manage an open access author’s fund at my library and we have funded articles at price points ranging from $2000 down to $175ish. And in case you’re wondering the $175ish article was published in Acta Crystallographica E which is not exactly a vanity publication.

    That being said I do think that a serious rethink of the journal system and the reward structure in academia is in order.

    For one thought on how the journal system could work see

  3. Stephen Curry (@Stephen_Curry) Says:

    I think Jan has covered almost all the points that I wanted to make (and some that hadn’t thought of) but just to add my tuppence worth:

    Objection 1 – it’s true that OA has created a market for some shady operators who really are offering vanity publishing but they are becoming well known for such dubious practices and are being shunned. You won’t find me submitting to Bentham, for example, in part because they were caught not doing proper peer review and in part because they emit such an incredible amount of spam. I think the community can do a good job of keeping the literature honest here.

    Objection 2 – Just to add that PLoS ONE has a no questions asked waiver for authors who cannot pay so, for this journal, there is no barrier to publication for hard-up authors. It seems entirely plausible to me that such a mechanism could work for other OA journals that work in the same way. I see it as a kind of progressive tax on richer authors/institutions.

    Objection 3 – This would be nice if it worked but I’m afraid the ’email the author route’ simply fails too many times. It will fail even more often in future as we scientists shuffle off our mortal coils. We need a system of access that works (and that, as Jan says, enables data and text mining so that we can extract maximum value from published research).

    Objection 4 – The economics are complex but to a first approximation I believe Jan is correct to point to current library subscriptions as a research cost. It may not be seen as such because funding comes via HEFCE (about £200m per year in the UK for journals and database access). Publication is integral to research and it’s costs should rightly be seen as part of the cost of research. This has the added advantage of making the costs transparent to the scientific community (they are anything but at the moment) and could provide useful market stimulus to drive costs down.

    There is a question about the additional cost that very research active institutions or nations would pay under OA versus a subs-based system since they produce more papers but this takes me back to the ‘publication is part of research’ argument. The Wellcome trust estimates the cost of publication is somewhere between 1 and 1.5% of research costs.

  4. James Says:

    I really don’t agree with your ideas here – for your Einstein point I think he would have quite happily published all his papers on ArXiV for free and not bothered with paying a journal.

    There’s a whole lot of activities going on at the moment in scientific publishing only some of which are paid for. Most importantly of the really difficult activities only the writing of the paper is generally paid, the reviewing is not. Publishing is now a fairly easily accessed and hence fairly valueless activity yet the publishers are the ones making all the money.

    Requirements of the academic publishing system:
    1) The paper’s author requires that the paper be made accessible quickly to as large an audience as possible for as long as possible.
    2) Their employers/funders often require some assessment is made of how good the paper is.
    3) The readers require that good checking of the paper occurs before publication so that it is readable and they can trust what they are reading, they may also like some kind of pointer towards better quality papers.

    It would seem to me that 1) can be achieved with such little cost by hosting on the internet that it doesn’t really matter who pays for it. 2) and 3) could be achieved by some kind of scoring system where all papers, reviews of papers and authors received a score.

    • James again Says:

      and the scores could for example be based on:
      1) how many citations each paper got and the scores of the papers they were cited by
      2) the number of useful reviews the author made and how quickly they made them
      3) how positive the paper’s reviews were, whether it was interesting enough to get a lot of reviews and views and the score of the reviewers
      4) reviews would be also be re-reviewable by the author and openly, non-anonymously reviewable once the paper was published, which would change their and the reviewer’s score. The paper would also continue to be reviewable after publication.
      5) some kind of balancing of the score for reviewing so that enough reviewing was done.

      however it worked sorting out a scoring system should be a lot easier than sorting out a search engine’s page ranking algorithm.

      • Jan Velterop Says:

        @ James again: it is worth checking out Utopia Documents ( which is a scientific PDF viewer, optimised for life sciences and biochemistry (although it works generically), and when PDFs of articles are opened (even informal ones, e.g. author manuscripts deposited in repositories), the article’s Altmetrics ( are automatically shown.
        Other metrics (e.g. Total Impact – and most likely PLoS Article Level Metrics) will soon be incorporated as well. Utopia Documents’ comment function also makes post-publication review very easy indeed, which was hitherto limited to HTML versions of articles.

      • protonsforbreakfast Says:

        Sorry for the delay in replying. That looks like a nice system, and I am happy to just publish in any system. Are you suggesting this format as a self-publishing alternative to publication by a journal?

      • protonsforbreakfast Says:

        This sort of makes sense if everyone uses it. So you are advocating self-publishing with a kind of open reviewing process?

      • Jan Velterop Says:

        Utopia Documents is not a publishing system, but a PDF reader that is made to bridge the differences with HTML where it concerns web connectivity. As a result, it has many features, including a very neat post-publication comment facility (which can be used for pre-publication peer review for manuscripts that are in PDF format, even PDFs created in wordprocessing software such as Word). I am indeed advocating wide usage of this remarkable software. It’s free, so there are no cost barriers.

    • protonsforbreakfast Says:

      Yes, reviewing is critical, and hard and time consuming.

      I agree with 1 2 and 3 though I am utterly sceptical of attempts to evaluate ‘how good’ a paper is.

      Mmmm. You don’t really address the issue of who pays. Somebody has to.

      FWIW I also do not consider ‘the internet’ to be a long term storage medium. I can go into the library at NPL and read the Proceedings of the Royal Society back to Newton. It is perfectly conceivable that the internet could ‘evaporate’. Storing multiple paper copies in diverse places has proved a pretty good method of storage over centuries rather than a decade.

      • James Says:

        if we are purely talking about the cost of putting something on the internet for others to be able to download the costs really are trivial.
        Let’s assume we are using an old fashioned single location.
        My last paper was 640 kB including diagrams etc. On a large scale cloud storage is available at ~10p/GB/month and downloads at ~10p/GB, both drop each year. If I was to pay £1 when I uploaded my paper to a paper hosting website that would more than cover hosting costs for the next 1344 years.
        Other people could download my paper 161 times for 1p. The most minimal amount of time spent doing something on mechanical turk would pay for plenty of downloads for a user. Personally I’d quite happily pay another £1 to cover the cost of the first 16000 downloads.
        But there’s equally no reason why a BitTorrent type approach couldn’t be used with the links hosted by a Pirate Bay type website funded by adverts, government grants or whatever. Then every person who reads the paper can potentially make a backup on their own computer.
        I’m sure mirrors would take it upon themselves to copy and host papers for free anyway in the same kind of way that wikipedia mirrors exist.

  5. Rosario Says:

    In developing countries all scientific journals are published by the universities and public research centers, and all of them are free even in paper or online. Why? because we have scientific development and we need to be read.

  6. James Says:

    this paper may be of interest:

  7. A couple of different aspects of accessibility. | …not that kind of psychologist Says:

    […] Protons for breakfast has some objections to Open Access.   […]

  8. A couple of different aspects of accessibility. « Åse Fixes Science Says:

    […] Protons for breakfast has some objections to Open Access.   […]

  9. Where to submit your paper? « squirreledthoughts Says:

    […] But really, what is a better option than submitting to Elsevier journals? I don’t really see the point of deliberately favouring one big publisher over another – Springer or Wiley journals are probably pretty similar to Elsevier in how much money they cash in per article. Open access journals sound like a pretty cool idea but my PhD consumables budget is limited and so I can’t really afford to go down the open access pay-yourself route (e.g. ~£855/article in PLoS One) and feel far too junior to be suggesting how my PI’s grant money gets spent, not to mention that open access has its own set of potential disadvantages. […]

  10. What kind of journal publishes nonsense? « Protons for Breakfast Blog Says:

    […] which means that if you pay them $500 – a low rate – they will publish your paper. As I mentioned previously, everything happens more smoothly if the quality barrier is as low as possible. The journal staff […]

  11. M. Says:

    Dear Michael,

    I like your article because it is exactly saying what I think about open access.
    Open access journals are run by commercial companies and they want/have to maximize their profit by the matter of our system. It was not the whish of the researchers who wanted free access to articles – instead it was a simple calculation. To charge each publishing author 1.5 K yields simply more profit than selling subscriptions to the university libraries.

    In my institution I have often contact with very young students. Interestingly all of them I asked so far like open access and more important: they argue open access saves the library. Usually I answer that if all articles are available by open access no one needs a library anymore.

    However I feel that it is more the expectation young students have that their supervisor will anyway publish the papers and therefore there will also be enough money to pay for publication.
    Those young students maybe still have to learn on how science really works but I wonder that also the most postdocs I know argue very similiar. I think no one of them has ever submitted a paper independently – only if you have to decide whether to pay or not you really knows what open access means.

    Best Regards,

  12. Fördelar och nackdelar med Open Access Says:

    […] [10] Open access is not such a smart idea, från bloggen ”protons for breakfast”. Publicerad den 25 april, 2012 klockan 6:00, läst den 30 december 2013. Länk: […]

  13. Topic 5: Advantages and disadvantages to a content producer of making their materials freely available online | Sylvia. Says:

    […] Personal. (2012, April 25). Retrieved December 9, 2014, from […]

  14. T5: Open Access to Online Materials | DiliniSene Says:

    […] (5), it is not likely that additional charges are contributing to any running costs. Yes, there are problems witth the current system of academic publications (it’s expensive which makes access […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: