Recently I've written a couple of entries on Journals and open access. In the latest WIRED, there's a good article on Harold Varmus, and his quest to shake up the biomedical journals.
From the article:
Varmus is the most visible character in the movement to free the scientific world of its figurative corks: scholarly journals that restrict the flow of information by charging often hefty subscription prices for access to their content.
...
He calmly lays out his campaign. For centuries, journals have been the means both of disseminating scientific knowledge and building scientific careers. Accordingly, the journals atop the hierarchy draw the highest-quality submissions, which reinforces their lofty reputations, which in turn enhances the status of the scientists who publish there. This positive feedback loop puts the power in the hands of the journals, even though their existence depends entirely on the scientists who write, edit, and serve as reviewers, usually without compensation.
Meanwhile, their colleagues can gain access only through subscriptions that their institutions pay for, sometimes dearly. (A yearly subscription to Brain Research, for instance, costs more than $20,000.) Worse, most of the public - scientists in developing countries, faculty and students in underfunded colleges, high schoolers, patients - have no access at all, even though taxes fund the government grants that support much of the research. Varmus asks: Shouldn't this ancient system have changed with the Internet, which allows information to be disseminated cheaply and immediately searched, mined, archived, reviewed, and improved?
- Log in to post comments
Yes, yes, yes, yes, yes, yes...
I think open access to top level work is THE central issue in science. Nothing holds science back more than limited access to the peer-reviewed database.
Apart from anything else, the fact that we the public pay for most scientific research means we should also have unfettered access to the results of any publicly funded research.
The idea of (at least part of) the publication costs being built into grants is a good one, which could largely solve the cost problem for the journals.
Those who describe open access to peer-review literature as some form of 'socialism' are, well, narrow-minded idiots who shouldn't be in science.
Yes but......
There are two different issues here. The first is the shape of journals in the electronic era. The second is the price of journal subscriptions and the role of for-profit publishers. Since the price per page for society run journals is about 10 percent of the price for for-profit journals, a large part of the dissatisfaction over cost has to do with the reliance on commerical publishers. I'm not sure how you get rid of them. They have a very small market share in astronomy, so the question doesn't really come up for us.
As long as we have journals we're going to have costs. Sometimes these are covered through page charges, sometimes through subscriptions. Some journals use a mix of the two. Open access, in the extreme case, removes the possibility of using subscriptions and puts the burden entirely on the authors. Even for society run journals this can be an unreasonable burden. Delayed open access, e.g. starting a year after publication, preserves subscription income, but does little to eliminate the high subscription costs for journals run by commercial publishers.
Taking advantage of the possibilities afforded by electronic publication should lower the costs for society journals by a significant factor (2?), provided we can agree to drop the goal of publishing hard copy versions, but it won't eliminate these problems.
You could run a very cheap journal (maybe down by a factor of 5 or more) by eliminating copy-editors and all other vestiges of post-acceptance manuscript processing. You might not like the result, but it would probably accomplish the minimum goals of a refereed journal.
You can run a non-refereed journal for nothing. I'd argue that it would be worth the cost.
Oh, my goodness! It is so clear and simple as Mr. Ethan stated in his closing statement. A feature on which the information-controlling elite representated by Mr. Varmus will not relent.
The efficient and free dissemination of information is likely only to change with the reformation of the sociopathological arbitrary, wasteful, and costly system of censorship and control of what information gets disseminated or is "valid" currently going under the guise of "peer review."
It all seems so silly and wasteful. Information should be peer reviewed by the large scale and deliberate evaluation of readers after it is published and available to the public, not before.
[Despite everything, I believe that people are really good at heart--Anne Frank]
Polly Anna, I think you missed Ethan's point -- that being that a non-refereed journal would cost nothing and be worth the same -- nothing. There are certainly problems with peer review but I think both scientists and, to a greater extent, the public, would not be happy with the outcome if peer review was dismissed. One of the primary functions of peer review is to double check that the statistical tests are correct, that the controls are appropriate and that the data is worthy of publication in the forum where it has been submitted. While all trained scientists should be capable of doing this on their own, sometimes one cannot see the forest through the trees of their own research and a unbiased referee is an outstanding way to re-examine the claims and make sure the controls and data analysis are appropriate to the conclusions.
Another problem I see with throwing out peer review is the breakdown of the current structure of manuscripts. One of the great things about a manuscript structure is that they generally tell a story with an appropriate context. This allows anyone with some training to sit down with any paper and figure out what is going on and what it means with some guidance from the authors. If there is no peer review and total open access, presumably you would no longer have a "journal" structure. While there would still be an incentive for putting out data in a manuscript form, I wonder how long that would last since it is quite time consuming and tedious (although it is clearly to all of our advantage). If data was disseminated piece-meal I'm afraid that only experts in the feild would understand what it meant and be able to place it in its context.
Dr. Varmus has achieved 2 great things with his PLOS Biology journals. 1) They are truly open access and 2) the peer review system in place there functions only as a check to make sure the papers are scientifically valid. Despite all the editorial bickering going on at the BMC journals, they have also instituted the same policies and many of them appear to be doing quite well.
FInally, i think peer review could be largely fixed (as Alex has pointed out before, I think) by simply removing author names. Sure some people would know who the authors are, but there would still be an element of doubt. As scientists we try to remove bias from the process as much as possible. Why not do it for the thing we all depend on for our livelihood?