Tucked in the spending bill passed by Congress is a "provision that would give the public free access to the results of federally funded biomedical research" (Washington Post, December 21, 2007) "The idea is that taxpayers, who have already paid for the research, should not have to subscribe to expensive journals to read about the results." The influence of "patient advocacy groups seeking easier access to the latest medical findings and supported by libraries whose budgets have had trouble keeping up with rising journal subscription costs" is quite obvious. Because a voluntary system proved unsuccessful (only 5% of researchers participated), scientists who receive funding from the NIH (National Institutes of Health) will have to submit their papers to a database once they are accepted for publication, and will see their findings available online for free within twelve months. While the bill focuses on NIH, it would make sense for other governmental funding agencies such as NSF (National Science Foundation) to align themselves with this precedent - taxpayer-funded research is taxpayer-funded research, even in topics less fashionable than biology or medicine.
In this day and age, when many researchers post their preprints on their website, and given that submissions to the vast majority of journals are free, published authors receive no royalties and reviewers do not get paid for their work, it is hard to see the value in the print publication of scientific results (why should journals get paid for advances they didn't fund and that can be disseminated without them?) - except for one factor, the prestige of the journal these results are accepted in, the only proxy currently available for their quality (and a flawed one at that, since reviewers must be well-versed in the topic to be able to judge the novelty of the results; in narrow, cutting-edge fields, they might well be competitors with a built-in interest in delaying acceptance of the paper - especially in experimental disciplines such as chemistry and biology where the trick is more often than not to figure out which experiment to perform). The perceived prestige of publications plays a significant role in tenure and promotion decisions in academia, and said journals are ranked every few years in terms of some "impact factor" more obscure than the U.S.News college rankings. But is the need to determine the quality of scientific publications worth the cottage industry of journals that charge libraries several hundreds of dollars a year for access to papers they did not commission and publish in print with long lead times after acceptance (typically over twelve months, because of backlogs)? Isn't there a more efficient way to disseminate novel findings?
I will say that the advantage of publishing papers both in print and online eludes me. Most scientists identify the papers they are interested in by searching online databases, and the print edition seems not only redundant but a waste of paper that drives costs for libraries (at the same time, though, if everything goes digital and people can consult databases from their office, libraries will become repositories of dusty outdated thirty-year-old textbooks, a fate their staff might want to avoid). It seems that online journals would help speed publication (no need to schedule the paper in a specific issue, no need to postpone publication if a future issue is already full) and keep costs down. I would be curious to see whether the disappearance of the bottleneck (paper, i.e., available pages, in the print issue) would lead to more papers being accepted if hard decisions no longer need to be made, in which case the move to digital archives would threaten the all-so-precious rankings, although some would certainly welcome a move to an Amazon.com-like system of rating papers online - the rankings at least in my field do lead to some puzzling results, and ranking papers themselves rather than publications might yield deeper insights into quality.
Does (belated) publication in a free online database mean the end of scientific publishing altogether? People will always need experts to tell them whether a result is noteworthy. Publication in a top journal implies precisely that; a free database should also provide that piece of information to help laypeople sift quickly through hundreds of papers and focus on the important ones, or at least let users provide a rating to distinguish the truly groundbreaking from the best forgotten. (The idea isn't mine; my then PhD advisor, Prof. Dimitris Bertsimas from MIT, came up with it a while back.) But once readers know where the paper was accepted and can access its contents with ease, a scientific publication has less incentive to keep operating, and while the end of for-profit publishing might not be a bad thing, its end before another quality-assessment is put in place certainly would be. Advocacy groups are kidding themselves if they think having earlier access to scientific papers on NIH-funded research will help them finding a cure for their loved ones faster, but I wouldn't want to be the biology professor, interrupted in his research by unwanted phone calls, who has to explain that to frantic relatives of terminal-stage patients.
I wish a period of Christmas full of serenity and joy and a happy new year to Aurelie and all the readers of this amazing blog and Not Only Bridges.
Posted by: xm carreira | December 22, 2007 at 06:57 AM