Show me the data!
Header

U2in
Just a quick post.

I happened to see @wisealic Tweet about her “new Atira/Pure colleagues” yesterday. I didn’t know what Atira was, but I’d heard of PURE.

I googled it to find out more… and soon found the official Elsevier press release , dated August 15, 2012 (so this isn’t really new news). But combined with recent rumours it does worry me. Elsevier own perhaps a fifth of the academic literature, whatever the true figure it’s a significant share. Despite the research that went into most of those papers being publicly or charitably-funded, Elsevier now rent access to this work back to us (the world) for vast sums of money each and every year.

Not to mention the fake journals they published, the arms dealings their parent company (Reed Elsevier) was involved in, their initial support for the RWA (since withdrawn), the megabundling of journals, the non-provision of open bibliographic metadata (even NPG release this!), the obscene profit margins (and to be fair they’re not the only corporate publisher making a killing here by selling freely provided academic work),  there are 1001 reasons why –  this isn’t an exhaustive list of all the evils…

So Elsevier are not a well-loved company in academia at the moment – more than 13,000 people have signed a boycott of them.

There are rumours that Elsevier are in talks to buy Mendeley at the moment. And Atira/PURE now part of the Elsevier (Umbrella?) corporation are I think the exclusive(?) providers of the research information ‘management’ systems that the UK will be using for it’s next Research Evaluation Framework (REF formerly RAE) exercise in 2014.

So… Elsevier own a significant portion of our papers,  and they may soon own a significant chunk of the bibliographic metadata stored by academics (Mendeley data) and all the commercial insight and advantage that gives, AND they own the company that is managing the data that evaluates UK academics and more round the world no doubt.

I do wonder if there isn’t a significant conflict of interest if thousands of UK academics have publicly boycotted Elsevier and now their academic work is going to be evaluated by… Elsevier. Academic jobs thoroughly depend on the results of these evaluations as I understand it, and heads will roll if the results at an institution are below expectations.

From a purely business perspective many financial analysts would rightly applaud these acquisitions as “good business moves” (good for profits no doubt). But from an ethical standpoint? Elsevier now seem to have a worrying empire of services built around academia and a significant amount of data which presumably they can pool together from each of these different services to gain additional insight? They also have a very poor record when it comes to providing open data. Why are we still giving them our data so easily – they’re only going to rent it back to us at a later date?

To me it’s clear, we’re giving up far too much of our data to this company and they do not have our best interests at heart – shareholder profits are by definition their primary goal. They have a sizeable monopoly on academic data in all it’s forms which they can and do leverage and I suspect we’re going to be made to pay for this mistake in the future as we have with hugely inflated journal subscription prices.

Is it just me that’s worried?

The Ecological Society of America (ESA) would like your input on how to expand access to their publications and what they should do if *gasp* the USA also mandates some form of public or open access …like the rest of the world seems to be doing at the moment.

The official call is here in this new free to access ESA publication (at the end):

Collins, S., Goldberg, D., Schimel, J., and McCarter, K. 2013. ESA and scientific Publishing—Past, present, and pathways to the futureBulletin of the Ecological Society of America 94:4-11.

You should probably read it all, so you understand their position and their misgivings before you email them with your ideas at: pubsfeedback@esa.org

 

Well done ESA. It’s nice to know they are aware of the inevitable changes that are happening in the world of academic publishing. They haven’t exactly been to receptive to the idea of Open Access before but now they seem to be acknowledging that it might be thrust upon them whether they like it or not and so need to prepare for it. I only wish all learned societies were doing this (I know we at the Systematics Association have plans, and that the Geological Society of London have a working group on this).

 

Here’s the email that I sent to them on Wednesday 23rd January (UK time). Proof, just in case they pretend they didn’t receive it *wink*

(N.B. I’ve recycled much of this from my House of Lords inquiry submission. Why not? Takes a lot of effort to write a detailed letter of support for Open Access! I’ll be damned sure to get some usage & re-usage out of it!)

 

 

Dear Ecological Society of America,

I read your special report ESA and Scientific Publishing—Past, Present, and Pathways to the Future with great interest. I wholeheartedly agree that the “world of scientific publishing is undergoing dramatic changes” at the moment – the internet clearly allows for extremely lowcost, efficient and open dissemination of research.

Currently there are huge inequalities in access to scholarly outputs (not just papers, but data & software too). My research library at the University of Bath can only afford to subscribe to so many subscription access journals – very far from all of them. But for myself and my colleagues to do high-quality, high-impact, definitive research we frequently need access to materials we don’t have either free/Open Access, or quick paid-subscription access to. In these cases myself and colleagues often spend hugely-wasteful lengths of time trying to get copies of these must read materials that are buried behind paywalls we can’t unlock.

The alternative options for access to paywall-restricted papers are poor and inefficient; inter-library loans can take days or weeks. Relatively few researchers currently post full-text self-archived copies of their own work in ‘green’ online repositories (although perhaps more might do so in the future). Electronic inter-library loans from the British Library can only be printed-once – if an error occurs during printing – tough luck, you’ll only ever have half a print version.

Sympathetic colleagues at different institutions with different journal access rights pass each other PDFs all the time – technically this is copyright infringement. Yet these small acts of academic copyright infringement are rampant online if you know where to look and are often the only way to sensibly and efficiently get research done. Buying additional legal access is simply not affordable nor desirable at the outrageous prices often offered – and sometimes only upon inspection of the fulltext does one find that the paper isn’t actually of use and can be discarded.

Many different peer-reviewed papers have shown that Open Access research has a higher citation rate than its paywall-protected ‘Closed Access’ counter-parts [e.g. 1-8]. Making ESA published research 100% Open Access would reasonably therefore confer some of this effect and increase the already impressive global impact of this research.

As you know the UK is far from alone in strongly pursuing Open Access means of research dissemination. The NIH Public Access mandate requires that all NIH-funded research publications are accessible to the public (world-wide) via the PubMed Central repository no later than 12 months after publication. In Australia, both NHMRC & ARC have Open Access policies in place. In fact if one looks closely enough one will see a litany of national research funders that already have open access mandates in place ArgentinaDenmarkAustriaBelgium, as well as innumerable policies at the university/institution level e.g. the Howard Hughes Medical Institute , Wellcome Trust, and even my own institution – the University of Bath (important to mention, because not all UK university research is funded by RCUK).

In particular I think we should note the way in which the SciELO Network has provided sustainable free access to over a thousand South American, Latin American, and (more recently) African research journals via the internet. It is ethically awkward that ‘they’ provide access to so much of ‘their’ research to us for free whilst we often charge them for access to ‘our’ research (many institutions do NOT receive charitably given access via HINARI ). This is an asymmetrical access imbalance that sorely needs to be corrected.

 

Learned Societies and Open Access

Learned societies heavily-reliant on subscription journal income and concerned with how public/open access policies may affect them should closely examine the workings of other societies that have successfully operated open access journals for many years. West and colleagues [9] provide robust data showing hundreds of society-operated gold Open Access journals with good citation impact at either no-cost to authors, or for a usually reasonable APC (article processing charge).

Good examples include the Journal of Economic Perspectives (of the American Economic Association) – not only do they charge nothing to authors (APC=$0) and provide free access to readers, but also Thomson Reuters Journal Citation Reports (JCR) ranks this as the 5th best journal in Economics out of 321 listed. It is influential and extremely well cited.

The journal Acta Veterinaria Scandinavica is a remarkable success story of society journals (it’s the official journal of the Veterinary Associations of the Nordic Countries). From 2000 to 2005 it was subscription-access only and was dwindling in impact and citations. In 2006 they changed to Open Access publishing with BioMed Central and now enjoy significantly increased impact and citations for the research published there.A plot of the Impact Factor of the journal Acta Veterinaria Scandinavica over time, showing a marked increase after switching to Open Access publishing. Source. Author: BioMed Central. Image licensed under the Creative Commons Attribution 3.0 Unported license

The European Geological Society (EGU) publishes 14 different gold Open Access journals with the help of Copernicus Publishing. One of these in particular – Atmospheric Chemistry and Physics has been hugely successful and through high citation rate is now ranked the 2nd best journal of 71 in the category “Meterology & Atmospheric Sciences” in Thomson Reuters JCR. It happily publishes articles using the Creative Commons Attribution Licence (CC BY) and charges a fair, variable APC that is cheaper for those who submit manuscripts in LaTeX form – reflecting the ease of which it is to convert such manuscripts into publishable forms. Microsoft Word submissions require more processing and thus they charge more (reflecting real cost). It is commendable that they expose, and make avoidable some of the labor costs of typesetting this way.

Furthermore, I’d bet there are many different societies operating subscription access journals that already allow self-archiving of published works so that they’d be compliant with the ‘Green’ OA route which the RCUK policy also allows. This would seem to me to be a fairly pain-free way of complying with the policy should ESA wish to do so via this route.

Overall, I think it would be fairer for all societies to publish associated journals in an Open Access manner – whilst clearly delivering on their core mission(s) of educating the world (not just a few subscribers) about their subject. Relying on denying access to research via paywalls to provide surplus income with which to spend on outreach and other activities that further the society mission, seem to me like a very convoluted justification and an inefficient way of achieving outreach goals. Put simply, Open Access very clearly fulfils many of the core purposes of learned societies and provides an open platform with which to build outreach around.

 

Finally, I would like to respond to some specific points that you mentioned in ESA and Scientific Publishing—Past, Present, and Pathways to the Future. 

  • “Will publishers need to invest heavily in their online platforms to meet gold requirements?”

Categorically, no. The current system of maintaining a sophisticated paywall, with login access only for paid-subscribers must be far more expensive to maintain and police than a simple, un-paywalled system whereby anyone can download articles. You already publish Ecosphere in a free to access manner which clearly shows you have the technology already in place to do this, so why suggest it would take heavy investment? Furthermore for societies that lack establish open access publishing systems there are plenty of cost-free (software-wise) robust solutions like Open Journal Systems that is already used successfully by over 11,000 journals (both Open Access & subscription journals!).

  • “Most publishers, including ESA, currently operate under a Creative Commons CC-BY-NC license for open-access publications…”

This is simply not true. Relatively few publishers and journals use this licence e.g. Jornal de Pediatria (a Brazilian journal). In fact, the majority of Open Access journals listed in Thompson Reuters JCR use the Creative Commons Attribution Licence (CC BY). Don’t believe me? Look at the data yourselves here. It’s the licence that BMC, Springer Open, PLoS, Hindawi, MDPI, Versita, Frontiers, Copernicus, Ubiquity Press, Pensoft, American Physical Society, and some Nature Publishing Group, Wiley & Sage open access journals use. So by counting publisher, journal or article-volume it’s definitely the most common Creative Commons licence used to publish scientific research. It’s common for very good reasons, not least that the non-commercial NC-clause can obstruct textmining analyses, and prevents the content from being re-used in Wikipedia.

  • You use an argument that ‘the “shelf life” of ecology research tends to be much longer than for medically oriented sciences

Whilst I don’t wish to disagree with you on this, I think you need not compare yourself to such a niche area of STM publishing. Take for example Palaeontology. I collected data recently to show that the mean age of a cited paper in a typical palaeontology article was >18 years! Yet in palaeontology there are plenty of successful high-impact open access journals and many which allow the green route of open access after a relatively short embargo period. If short (6 or 12 month) embargo periods don’t affect the income of subscription access palaeontology journals, why would it cause harm to ESA journals to allow this? I feel you fear something that won’t actually happen.

  • I strongly doubt that if you allowed a ‘green’ friendly route to Open Access, with a 6 month embargo as allowed by the RCUK policy, that you’d lose much subscription revenue.

Statistics from the Romeo/SHERPA database that tracks green OA policies shows that 60% of journals allow immediate self-archiving of the full-text of research papers, with a further 27% permitting the submitted version (pre-print) to be archived immediately. Only 13% of journals do not allow immediate archiving. There remains little convincing evidence that short embargo periods seriously harm library subscription revenue.

So if I were ESA, I’d probably look into the green OA route as a relatively pain-free / hassle-free way of expanding public access to research.

 

Regards,

 

Ross Mounce,

PhD Student at the University of Bath  & Open Knowledge Foundation Panton Fellow

http://about.me/rossmounce

 

References

 

1. Lawrence, S. 2001. Free online availability substantially increases a paper’s impact. Nature 411:521 http://dx.doi.org/10.1038/35079151

2. Xia, J. and Nakanishi, K. 2012. Self-selection and the citation advantage of open access articles. Online Information Review 36:40-51.http://www.emeraldinsight.com/journals.htm?articleid=17004555&show=html  [the OA citation advantage is more pronounced for ‘smaller’ journals]

3. Xia, J., Myers, R. L., and Wilhoite, S. K. 2011. Multiple open access availability and citation impact. Journal of Information Science 37:19-28.http://dx.doi.org/10.1177/0165551510389358 [More copies available in different places, more citations…]

4. Riera, M. and Aibar, E. 2012. Does open access publishing increase the impact of scientific articles? an empirical study in the field of intensive care medicine. Medicina intensiva / Sociedad Espanola de Medicina Intensiva y Unidades Coronarias.http://dx.doi.org/10.1016/j.medin.2012.04.002

5. Norris, M., Oppenheim, C., and Rowland, F. 2008. The citation advantage of open-access articles. J. Am. Soc. Inf. Sci. 59:1963-1972.http://dx.doi.org/10.1002/asi.20898

6. Eysenbach, G. 2006. Citation advantage of open access articles. PLoS Biol 4:e157+. http://dx.doi.org/10.1371/journal.pbio.0040157

7. Hajjem, C., Harnad, S., and Gingras, Y. 2006. Ten-Year Cross-Disciplinary comparison of the growth of open access and how it increases research citation impact. http://arxiv.org/abs/cs.DL/0606079

8. Gargouri, Y., Hajjem, C., Larivière, V., Gingras, Y., Carr, L., Brody, T., and Harnad, S. 2010. Self-Selected or mandated, open access increases citation impact for higher quality research. PLoS ONE 5:e13636+. http://dx.doi.org/10.1371/journal.pone.0013636

9. West, J., Bergstrom, T. and Bergstrom, C. T. 2013. Cost-effectiveness of open access publications

 

Here’s my submission for the House of Lords inquiry. I rather ran out of steam writing it so you’ll see it tails off towards the end. There’s probably loads of things I should mention too. But alas, I have lots of other work to be getting on with right now. Ironically, I highlight the excellent journal Impact Factor‘s of some OA journals. Please forgive me for those sins! So here it is:

 

17/01/2012 Author: Ross Mounce, final year PhD Student at University of Bath & Open Knowledge Foundation Panton Fellow email: rcpm20@bath.ac.uk

 

This submission is an individual contribution but I think it may be indicative of the opinion of many in the scientific research community. Of particular relevance to this inquiry I should state my research funding is from BBSRC, I am engaged in content mining research (which is commonly hampered by copyright/legal issues with respect to non-Open Access research), and I am a council member of The Systematics Association (a UK-based learned society that publishes academic works with CUP).

 

Background

 

  1. On the whole I was extremely pleased when the Finch Report came out and even more so when RCUK announced it was going to implement most if not all of the recommendations. I, and most of my colleagues strongly believe that taxpayer-funded research such as that given out by RCUK should be made openly available to everyone in the world to read and to use for whatever purpose (Open Access).
  2. Currently there are huge inequalities in access to scholarly outputs (not just papers, but data & software too). My research library at the University of Bath can only afford to subscribe to so many subscription access journals – very far from all of them. But for myself and my colleagues to do high-quality, high-impact, definitive research we frequently need access to materials we don’t have either free/Open Access, or quick paid-subscription access to. In these cases myself and colleagues often spend hugely-wasteful lengths of time trying to get copies of these must read materials that are buried behind paywalls we can’t unlock.
  3. The alternative options for access to paywall-restricted papers are poor and inefficient; inter-library loans can take days or weeks. Relatively few researchers currently post full-text self-archived copies of their own work in ‘green’ online repositories (although perhaps more might do so in the future). Electronic inter-library loans from the British Library can only be printed-once – if an error occurs during printing – tough luck, you’ll only ever have half a print version.
  4. Sympathetic colleagues at different institutions with different journal access rights pass each other PDFs all the time – technically this is copyright infringement – we have a system that appears to criminalise attempts to do comprehensive and diligent research. Yet these small acts of academic copyright infringement are rampant online if you know where to look and are often the only way to sensibly and efficiently get research done. Buying additional legal access is simply not affordable nor desirable at the outrageous prices often offered – and sometimes only upon inspection of the fulltext does one find that the paper isn’t actually of use and can be discarded.
  5. Many different peer-reviewed papers have shown that Open Access research has a higher citation rate than its paywall-protected ‘Closed Access’ counter-parts [e.g. 1-8]. Making RCUK research 100% Open Access should reasonably therefore confer some of this effect on our research and increase our already impressive global impact, particularly if we are one of the first big research nations to embrace this, rather than the last.
  6. But the UK is far from alone in strongly pursuing Open Access means of research dissemination. The NIH Public Access mandate requires that all NIH-funded research publications are accessible to the public (world-wide) via the PubMed Central repository no later than 12 months after publication. In Australia, both NHMRC & ARC have Open Access policies in place. In fact if one looks closely enough one will see a litany of national research funders that already have open access mandates in place Argentina, Denmark, Austria, Belgium, as well as innumerable policies at the university/institution level e.g. the Howard Hughes Medical Institute , Wellcome Trust, and even my own institution – the University of Bath (important to mention, because not all UK university research is funded by RCUK).
  7. In particular I think we should note the way in which the SciELO Network has provided sustainable free access to over a thousand South American, Latin American, and (more recently) African research journals via the internet. It is ethically awkward that ‘they’ provide access to so much of ‘their’ research to us for free whilst we often charge them for access to ‘our’ research (many institutions do NOT receive charitably given access via HINARI ). This is an asymmetrical access imbalance that sorely needs to be corrected.

 

On Learned Societies

 

  1. Learned societies heavily-reliant on subscription journal income and concerned with how the RCUK policy may affect them should closely examine the workings of other societies that have successfully operated open access journals for many years. West and colleagues [9] provide robust data showing hundreds of society-operated gold Open Access journals with good citation impact at either no-cost to authors, or for a usually reasonable APC.
  2. Good examples include the Journal of Economic Perspectives (of the American Economic Association) – not only do they charge nothing to authors (APC=0) and provide free access to readers, but also Thomson Reuters Journal Citation Reports (JCR) ranks this as the 5th best journal in Economics out of 321 listed. It is influential and extremely well cited.
  3. The journal Acta Veterinaria Scandinavica is a remarkable success story of society journals (it’s the official journal of the Veterinary Associations of the Nordic Countries). From 2000 to 2005 it was subscription-access only and was dwindling in impact and citations. In 2006 they changed to Open Access publishing with BioMed Central and now enjoy significantly increased impact and citations for the research published there.A plot of the Impact Factor of the journal Acta Veterinaria Scandinavica over time, showing a marked increase after switching to Open Access publishing. Source. Author: BioMed Central. Image licensed under the Creative Commons Attribution 3.0 Unported license 
  4. The European Geological Society (EGU) publishes 14 different gold Open Access journals with the help of Copernicus Publishing. One of these in particular – Atmospheric Chemistry and Physics has been hugely successful and through high citation rate is now ranked the 2nd best journal of 71 in the category “Meterology & Atmospheric Sciences” in Thomson Reuters JCR. It happily publishes articles using the Creative Commons Attribution Licence (CC BY) and charges a fair, variable APC that is cheaper for those who submit manuscripts in LaTeX form – reflecting the ease of which it is to convert such manuscripts into publishable forms. Microsoft Word submissions require more processing and thus they charge more. It is commendable that they expose, and make avoidable some of the effort costs of typesetting this way.
  5. Furthermore, I’d bet there are many different societies operating subscription access journals that already allow self-archiving of published works so that they’d be compliant with the Green OA route which the RCUK policy also allows (with additional leniency on the humanities, allowing a 12 month embargo). This would seem to me to be a fairly pain-free way of complying with the policy should they wish to (N.B. Learned societies are not obligated to comply with this policy, although you would think if it was a British society it might be in their best interests. It is the researchers that must comply).
  6. I am concerned for some UK learned societies that from their annual financial reports seem to indicate they are rather reliant on subscription-journal income to support their societies financially. I am not privy to the exact details of whether society subscription-journal income is ‘ringfenced’ away from supporting the other activities & perks of a societies’ membership. I hope it is. Otherwise I worry that perhaps some learned societies maybe using the surplus from the subscription-access journal income (paid for by libraries/institutions/universities world-wide) and spending this surplus on personal society member-only perks e.g. a free hardcopy paper newsletter only delivered to personal members. I have examined annual report accounts of some learned society accounts myself and find that where the money/surplus goes to be rather opaque in some cases.
  7. It appears that many societies have been operating a consistent and healthy surplus from their subscription-access journals and using this surplus to expand their outreach activities and member perks – free pens, paper, mugs, USB sticks and heavily discounted student memberships. I myself have greedily taken many of these membership benefits, and know that I have received goods and services that far exceed the cost of the small, hugely subsidized membership fee I paid. All this would be okay if it was only members paying for other (younger) members – self-sustainability. But I am increasingly concerned about the asymmetry of fees and benefits provided by some learned societies. Surely a significant portion of journal subscription income is from institutional subscriber agreements? Institutions are very rarely members of learned societies, and institutionally the only benefit they get from these fees paid is institutional access to subscription-only society journals. Yet the surplus from subscription income at societies doesn’t seem to be given back except to members through perks and the organisation of outreach events and such.
  8. Therefore I think it would be fairer for a society to publish any associated journals in an Open Access manner and concentrate on being financially self-sustaining – whilst clearly delivering on their core mission(s) of educating the world about their subject. Relying on denying access to research via paywalls to provide surplus income with which to spend on outreach to further their mission, seems like a very convoluted argument and an inefficient way of achieving their aims. Put simply, Open Access very clearly fulfils many of the core purposes of learned societies and provides an open platform with which to build outreach around.

 

Arrangements for APC funds

 

  1. As I’m sure many will cite, most gold open access journals listed in the Directory of Open Access Journals (DOAJ) are fee free. They do not charge an APC. Of those that do, the average APC is just $906 (Solomon & Bjork, 2012). There is no strong relationship between the APC cost of gold open access journals and their article level impact [9]. Intuitively this makes sense – if I submitted my work to Nature, or I submitted my work to the Panamanian Journal of Ichthyology (a fictional journal) the work, if published, would essentially be the same – journal ‘brand’ is just a label, it doesn’t change anything – especially not the quality of peer review. In terms of citations, solid evidence supports this intuition – since 1990 the relationship between Impact Factor (citations to a journal) and article-level citations has significantly weakened [10]. To put it another way – good research gets read and cited no matter where it’s published.
  2. I’m aware there are concerns in the Humanities and Social Sciences about Open Access and APCs. I don’t know why there aren’t more Open Access journals in these disciplines. There’s nothing technologically preventing a surfeit of new Open Access journals from forming. Good, well tested solutions like Open Journal Systems are free to implement (no software cost) and are used by over 11,000 journals world-wide. The implementation only needs bandwidth-cost support and the same human time/effort required to run a subscription access journal, which I’m sure institutions should be made willing to help with. Stuart Shieber gives an excellent description of how costs are managed at the Journal of Machine Learning Research. Here academics volunteer time, with the help of a little institutional support to produce a high-quality, high-impact peer-reviewed research journal that costs just $6.50 per paper to run.
  3. I would urge the House of Lords to look into how universities and libraries could be encouraged to help British academics create new, efficient, low-cost, peer-reviewed research journals. Martin Eve for one appears to have no trouble doing this. It need not even necessarily require additional cash-injection, just IT-support and the use of institutional bandwidth & servers to host Open Access journals. Willingness to try, rather than just moan about change is also required.
  4. Above all, academics in all areas need to consider and be made aware of the huge variety of open access publishing options available to them. The big commercial publisher brands may be the most well-known in some areas, and they spend significant marketing budgets on ensuring this. Unfortunately these commercial publishers also offer some of the most eye-wateringly expensive gold Open Access options. We need to incentivize and ensure a ‘value-for-money publishing’ mentality, and to discourage academics away from these expensive ‘hybrid’ OA options. It would be good to set a hard limit on the amount of cash that RCUK would be willing to pay for an APC for any one publication. Otherwise it might encourage some publishers to further indulge in price-gouging.
  5. I am glad that RCUK is supporting gold open access and green open access routes. I fail to see how green alone would work out in the end – it does not provide peer review. ‘Overlay’ peer-review services external to journal publishers operating on pre-print servers are a nice idea, but I’m not sure this model of publishing will gain traction or acceptance in academia, not for a while at least. Therefore to continue to build-on and support low-cost journals I think it is good that RCUK is encouraging the gold open access route.

 

Embargo periods

 

  1. I don’t have much to say about embargo periods. Only that I’ve seen some interesting arguments used against short embargo periods in the humanities e.g. history. One such argument used was that the ‘citation half-life’ was very long in History and therefore a short embargo period would harm this discipline more than in the sciences. Yet I know that in Palaeontology, the citation half-life of papers as you might imagine is also very long – yet there are few such concerns about embargo periods or the effect of Open Access in this discipline. I recently gathered data and found that the mean-age of cited papers in palaeontology is roughly >18 years. Therefore I don’t ‘buy’ this long-tail usage argument as it equally applies in other disciplines that appear to have no problem with open access, green or gold.

References

 

1. Lawrence, S. 2001. Free online availability substantially increases a paper’s impact. Nature 411:521 http://dx.doi.org/10.1038/35079151

2. Xia, J. and Nakanishi, K. 2012. Self-selection and the citation advantage of open access articles. Online Information Review 36:40-51.http://www.emeraldinsight.com/journals.htm?articleid=17004555&show=html  [the OA citation advantage is more pronounced for ‘smaller’ journals]

3. Xia, J., Myers, R. L., and Wilhoite, S. K. 2011. Multiple open access availability and citation impact. Journal of Information Science 37:19-28.http://dx.doi.org/10.1177/0165551510389358 [More copies available in different places, more citations…]

4. Riera, M. and Aibar, E. 2012. Does open access publishing increase the impact of scientific articles? an empirical study in the field of intensive care medicine. Medicina intensiva / Sociedad Espanola de Medicina Intensiva y Unidades Coronarias.http://dx.doi.org/10.1016/j.medin.2012.04.002

5. Norris, M., Oppenheim, C., and Rowland, F. 2008. The citation advantage of open-access articles. J. Am. Soc. Inf. Sci. 59:1963-1972.http://dx.doi.org/10.1002/asi.20898

6. Eysenbach, G. 2006. Citation advantage of open access articles. PLoS Biol 4:e157+. http://dx.doi.org/10.1371/journal.pbio.0040157

7. Hajjem, C., Harnad, S., and Gingras, Y. 2006. Ten-Year Cross-Disciplinary comparison of the growth of open access and how it increases research citation impact. http://arxiv.org/abs/cs.DL/0606079

8. Gargouri, Y., Hajjem, C., Larivière, V., Gingras, Y., Carr, L., Brody, T., and Harnad, S. 2010. Self-Selected or mandated, open access increases citation impact for higher quality research. PLoS ONE 5:e13636+. http://dx.doi.org/10.1371/journal.pone.0013636

9. West, J., Bergstrom, T. and Bergstrom, C. T. 2013. Cost-effectiveness of open access publications

10. Lozano, G. A. , Lariviere, V. and Gingras Y. 2012. The weakening relationship between the Impact Factor and papers’ citations in the digital age http://arxiv.org/abs/1205.4328v1

 

Anyone who knows me, knows I’m very passionate on the subject of data sharing in science, and after all the relevant conferences I’ve been to and research I’ve done – I don’t mind saying I’m fairly knowledgeable on the subject too.

It’s part of the reason I got this Panton Fellowship that has helped me develop my work and do what I want to do in pursuit of Open Data goals.

So when I saw this article come up on my RSS feeds – I thought great! It’s finally happening. The vertebrate palaeontology community is finally seeing the light – the absolute need to share research data associated with published papers (we’ll tackle pre-publication data sharing later, first things first…)!

Uhen, M. D., Barnosky, A. D., Bills, B., Blois, J., Carrano, M. T., Carrasco, M. A., Erickson, G. M., Eronen, J. T., Fortelius, M., Graham, R. W., Grimm, E. C., O'Leary, M. A., Mast, A., Piel, W. H., Polly, P. D., and Säilä, L. K. 2013.
From card catalogs to computers: databases in vertebrate paleontology. Journal of Vertebrate Paleontology 33:13-28.

2013-01-12-142813_1054x983_scrot

…and yet when I read the paper – it sorely disappointed me for a variety of reasons.

Choosing examples: bad choices & odd absences

Despite clear criteria given, I found the choice of databases reviewed to be an odd selection – for example they choose to include AHOB (Ancient Human Occupation of Britain) and write about it that:

“Access is restricted to project members during the life of the project, after which access will be publicly granted.”

This probably explains why then, that when I go to the database website – I can’t seem to get access to any of the purported data to be there!

AHOB
Screenshot of the login screen for AHOB. Try it yourself.

Yet apparently: “More than 250 publications have results from the AHOB project, all of which are recorded in the database.”

How many more publications will come out of this cosy little database before access will be publicly granted I wonder? I don’t think this is a good example of a research database as it doesn’t seem to publicly share any data.

Where’s Dryad?

Furthermore there are some really big, obvious, relevant databases it neglects to review, in particular Dryad – the only mention of which is that TreeBASE received “some support from Dryad” – with absolutely no mention anywhere that Dryad itself is a database with lots of vertebrate palaeontological data in it and likely to be a strongly important, long-lasting database in this area for the foreseeable future IMO! Even some data associated with an article in JVP itself is in Dryad! Although less prominently paleo-related figshare (with no less that 26 paleontology-related datasets there at the moment, TreeBASE has approximately as many!) might have been worth mentioning too.

Dryad has a partnership with The Paleontological Society and many evolutionary biology journals. Dryad even bought a promotional stand at last year’s Society of Vertebrate Paleontology annual meeting (the society that publishes the Journal of Vertebrate Paleontology) but as Richard Butler has pointed out to me on Twitter this article was submitted before that meeting. Still, it’s simply impossible that none of the 16 authors listed doesn’t know about Dryad. I find the non-inclusion of Dryad deeply suspicious and possibly political given it could ‘compete’ to store much of the data that some of the other reviewed databases do (it’s a broad generalist in the types of data it accepts).

Isn’t there a conflict of interest issue given that most of the authors of this paper are involved with at least one of the ‘reviewed’ (=advertised) databases in the paper? I see no mention of this conflict of interest anywhere in the paper. I dearly hope this paper was peer-reviewed – that it is an ‘invited article’ makes me wonder a bit about that…

The inclusion of Polyglot Paleontologist too, in the reviewed databases does also rather stretch the meaning of ‘data’ in the word database. Are translations of 434 different papers ‘data’? In the same way that TreeBASE or PaleoDB contain data? It’s a fantastic freely provided resource, no doubt – I mean no criticism of it – but is it data? I think not tbh.

Strong contenders for things that could/should have been cited but weren’t

WRT to Data Portals: rOpenSci provide great R interfaces for a wide variety of databases, including TreeBASE which was one of the ‘reviewed’ databases.

WRT to the History of databases section: I find it odd that they didn’t think to mention my own widely publicised and well-supported call for data archiving in palaeontology back in 2011. Nearly 200 palaeontologists signed in support of our ideas with some memorable quotes of support e.g. Brian Huber “This is the way of the future” , P J Wagner “I’ve been trying to get the Paleo Society to sign on with Dryad, but it’s been like slamming my head on jello…”

They could have explained why freely accessible databases/archives are so important a bit better in my opinion:
that ‘Data archiving is a good investment‘ (Piwowar et al, 2012),
that only 4% of phylogenetic data is currently archived and that it’s really useful data (Stoltzfus et al, 2012),
that Willingness to Share Research Data Is Related to the Strength of the Evidence and the Quality of Reporting of Statistical Results (Wicherts et al, 2011),
that the “data available upon request” system really doesn’t work (Wicherts et al, 2006)
the undesirable consquences of non-commercial clauses applied to biodiversity data (Hagedorn et al, 2011)

Odd wording

“…community approach, facilitated by the open access of the WWW and…”

sounds like something my dad would say about the interweb

“The CCL 3.0 license allows…”

a classic mistake – which CCL license?
In this case they mean the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 license, or CC BY-NC-SA for short. Calling it “Creative Commons License 3.0 (BY-NC-SA)” makes me wonder how familiar they are with licencing. Perhaps a sub-editor did this. And why they link specifically to the US version not the international unported license I do not know.

Data Citation: the Elephant in the Room?

Attribution is mentioned many times, and is vitally important to motivate people to share data. Yet the concept of citing data in countable ways or Data Citation isn’t explicitly mentioned once. Nor altmetrics for that matter.

This would have been an excellent opportunity – the start of a new year to encourage authors to actually cite data that they re-use from someone else so that those citations can be easily counted and contribute towards research evaluations, but alas no.

So what now?

So I like some of the message of this paper. But I don’t think it goes far enough, nor does a good job of it. Call me egotistical but I think I could do better and expand upon what I’ve written above.

If any journal editor happens to read this, and would like to commission an ‘invited article’, comment, or proper independent critical review of databases in vertebrate palaeontology / evolutionary biology please contact me. I think I could offer an interesting perspective.

PS I’m not going to write to the journal. I tried that with Nature and it took 6 months from submission for my comment to get published! It’s 2013 – if I’m going to do post-publication peer review – I’ll definitely be blogging it from now on, Rosie Redfield style!