Jake Young has written an excellent summary of a panel discussion he attended at the Society for Neuroscience meeting.
I encourage you to read the whole thing, as it presents a fascinating interplay of the forces at work in academic publishing. But if Jake's synopsis is too much for you, here's a quick summary of the issues involved:
- The current "market-based" scholarly publishing system is primarily paid for by governments: Researchers and libraries get grants, and the grants pay for subscriptions to journals.
- This system limits access: not everyone has access to libraries, and not all libraries have equal access. (To take one small example, the Davidson College library doesn't subscribe to the online edition of Science. That's because the institutional rate for the online version is very expensive, so they figure anyone who's really interested will just subscribe since Science is a really cheap journal to subscribe to on an individual basis. But $139 here, and $200 there, and pretty soon you're talking real money.)
- All the panelists agreed that since it's all government money anyway, the same research could be published under an "open access" model and made freely available to all.
- Not to belabor the point, but this would be a better system.
- The problem comes when you get down to details: the Public Library of Science, for example, already publishes several excellent journals in this way, but requires authors to pay a fee of up to $2,500 when the work is published, to cover editing, typesetting, and administrative costs. This could be a problem because not all researchers can afford the fee, so good research may go unpublished. (However, additional grants could be made available for these cases. Remember, it's all the same pile of money.)
- The bigger problem appears to be that publishers are unwilling to let go of their empires: Elsevier publishes over 1,500 journals, and makes a tidy profit doing so. Moving to a completely new model would be economically unstable to them, so they prefer to stick with the old system.
- Publishers appear incapable of understanding that there are other ways of distributing information. I'd like to see some wikipedia-style journals, some efforts to peer-review blogs, and some other, more efficient models developed. While peer review is a very good idea that shouldn't be abandoned, it's not necessarily the only way to distribute reliable research results.
Any other thoughts from CogDaily readers?
- Log in to post comments
It sounds nice in principle to go to a Wiki-style on-line open-submission, open-commentary peer-review system--which is my understanding of the xArchiv system used in physics. I don't think this would work in biomedical science, because there is orders-of-magnitude more research going on, and thus orders-of-magnitude more papers the system must handle.
One of the great advantages of the closed-submission, proprietary journal system is that the journal editorial process selects a *very* small fraction of the research performed for publication in a given journal. Depending on one's interests, this allows for a tremendous time and effort savings for the reader. This depends, of course, on trust in the closed editorial and peer-review process.
In my field, I need to read the tables of contents of only about 20 journals, and this gives me access to 99% of the papers that will be of interest to me. The other 1% I can be sure I will be exposed to via word-of-mouth from my colleagues.
One could imagine an open-access system whereby blog-type editorial Web sites are established with the purpose of replacing the filtering system that is currently implemented via closed editorial and peer review. This does not currently exist in a useful form.
--ttcc
The link doesn't work, because it has an extra " (double quote) at the end of the URL. Do please correct it.
It's not just Elsevier which is against open access journals. Professional societies, which ought to take care of the interests of their members, are no better. Case in point: the American Chemical Society.
I agree that the grantors should be hit up to include publication costs in the total amount they give to the researcher. It's certainly in their best interest - What's the point in paying for research that never gets published?
In fact, if the research is paid for with tax money, even 1%, it should be _required_ that the results are available to the public at no cost to the reader.
Wikipedias are really the best option for wide dissemination of quickly-updating information. If programmers develop open source search engines, they could very easily be improved upon and trained to look for exactly what the reader wants. A really useful engine (like Thomas) would have a feature to look for related articles outside of the reader's specialty, so that the researcher, with all their spare time, might learn what ideas in economics could give them new insight into protein folding.
Physioprof --
Good points. I'm not sure every solution will work for every discipline. I'd submit that biomedical sciences are excellent candidates for the PLOS system since so much of that research is supported by grants.
In psychology, where much good work is done outside of the grant system, it might be more difficult to implement, and some kind of wiki/blog solution might work better.
Abi,
I didn't mean to single out Elsevier in the post, that was just an example. With professional societies, I suspect that sometimes the journal subscription is a carrot to get people to join. If the journals were open-access, then those societies would have many fewer members. (And thanks for the note about the broken link. It's fixed.)
If Elsevier is worried about losing money from going open access, couldn't they make up the difference by selling more guns?
Thank you for this post. I come to it from Open Access News Blog by Peter Suber ( http://www.earlham.edu/~peters/fos/fosblog.html ).
At Neuroscience 2006 I had a teaching poster presentation on Open Access. Two years ago at Neuroscience 2004 I presented another session on OA co-authored with major US Open Access advocate Peter Suber. Two years ago SfN found our abstract hot enough to include its' lay language article in the Meeting Press Book. Both presentations (2006 and 2004) are available at this link as .PDF imprints of poster presentations and lay article: http://israelscholar.org/advocacy/
Sincerely,
Alexei Koudinov, MD, PhD
http://neurobiologyoflipids.org
http://dopingjournal.org
http://israelscholar.org
Some sort of Digg.com style system would work well too making the most interesting new research float to the top.
It is rather annoying how outdated many of the academic search engines are. I'm so used to getting exactly what I want via google or wikipedia right away, and then whenever I have to use a barebones search engine like infotrac it's excruciating. The results are often so meaningless that it feels like climbing through a junkyard.
Speaking of which, can I tell you how lame it is that PsycINFO is completely closed off? Why can't they at least display the abtracts like PubMED does?
I've expressed this idea several times in various places, but what I would like to see is an extension of arXiv to include peer review on the order of "you must review two papers, at least one of which has one or less previous reviews, before you can post another one". Since most researchers are reading papers in their field anyway, this would provided a good feedback system. Papers with low quality or questionable results will be pointed out by other authors, and people will do it because that's the cost of publishing.
In my opinion companies like Elsevier see that open access will gain "market share" in the years to come. But they have a field where they can contribute and make profits without the help of lock-in phenomena. That is the specialized databases and visualization tools (e.g. in medicine, biology, chemistry etc).
On the other hand open access is more favourable for the researches since their work is becoming widely available therefore more "contagious". When and if these open access papers become more successful more good papers will go that direction and publishers will also be forced to open up more than just abstracts. But have to keep in mind that they have a stock of few hundred years of good copyrighted material.
Physioprof - It it true that we have so far not developed the filtering tools but it is impossible to do so while we do not separate the certification and filtering process from the mere archiving of content. If we separate these two we can still use the current filtering (journal editorial boards and referees) but we can on top of this try out new systems (like blog reviews and tagging). This would be a controlled experiment that would not cost us anything. The quality and filtering would be maintained and we could compare the results of new filters with the existing process.
Most journals accept papers that were deposited in pre-print servers but most people are somehow afraid to submit their manuscripts to an archive before submitting to a journal.
Alternatively we could experiment with filtering using the published papers. In fact this is already happening in sites like citeulike, connotea and postgenomic.
You ain't just whistling Dixie, Colin. I'm an academic librarian and the clunkiness of some of the tools we're stuck working with drives me crazy. Part of the reason, again, is cost. A really good research database costs a lot of money, and even with that lot of money you often still get an interface that hearkens to the early days of the Web. I wish I knew what to do about it. Too often, libraries are stuck with what we can afford (and I work in a very small academic library--fortunately, most of the faculty I work with are aware of the harsh reality of cost).
I'm a fan of the open-access model, insofar as a model can be said to exist. Perhaps it's more accurate to say that I'm a fan of the concept. It makes things so much easier, not just for researchers, but for students who are often stuck with what their library has or can provide. We have an excellent interlibrary loan system where I work--requests often come in the same day they're made, an astoundingly good turnaround time--but it sure would be nice not to have to do that, or to make it more transparent to students.
To an extent, Elsevier is an easy target. Their journals are insanely expensive, as is pretty much everything else they publish (a bookstore buying manager I used to work for referred to them as "the root of all evil"). But no, they're not the only ones.
I like having openly accessible search indexes, such as PubMED. If my students could more readily see what's out there, they might hold the whole concept of searching the research literature in higher regard. Explaining that not everything is Google-able is now a basic component of a lot of my classes.