The Fall of the Scientific Wall

Democracy has reached a new frontier, and we’re not talking about the Berlin Wall. It’s a new decade and a new millennium, and yet another wall is crumbling—this time, not between countries, but in the domain of scientific research.

New Internet-based journals are challenging the status quo by publishing works that have not yet passed the usual, rigorous peer-review system, giving any cyber-citizen the power to appraise many novel scientific inquiries. And it’s all too easy to underestimate the potential for science this experiment brings.

The hermetic process of traditional peer-review does little to facilitate free intellectual dissemination. Scientific manuscripts are typically reviewed by a small number of anonymous experts, who have a throttlehold on whom and what deserves attention. In October 2000, three concerned biomedical scientists founded the Public Library of Science (PLoS), calling on fellow scientists to boycott journals that refused to make full-text papers part of the public domain within six months of publication. Yet although some scientific publishers conformed to the PLoS’s policy guidelines, most remained unresponsive.

Undeterred, PLoS continued its dedication to the principle of open access. This November awaits the launch of its first open peer-reviewed journal, PLoS ONE, which will welcome reports on primary research from any scientific discipline. Like other PLoS journals, authors will be assessed a fee in order to publish, but any paper will receive only cursory scrutiny for technical quality before it appears for open discussion by the online community.

Not everyone has been enthusiastic about such a visionary opportunity for scientific innovation, however. Critics suggest that this system allows the rabble to promote “junk science” and argue that scientists will have to wade through a hundred worthless papers to find only one Nobel Prize-winning gem. They also claim that rogue scientists could praise and criticize research in an unfair, un-objective way. But copious empirical evidence indicates that open online communities—including those dedicated to scientific research—have an incredible capacity to self-regulate.

Russian mathematician Grigori Perelman’s recent online publication of a proof of the Poincaré conjecture—a century-old question of fundamental importance in topology, the solution to which won Perelman a Fields Medal—is an iconic example of the unlimited possibilities in open online science. Perelman’s proof never underwent the trials of a typical peer-review process, but expert mathematicians have scrutinized his work to such an extent that it is widely considered to be correct. And this success is far from the exception to the rule. The database of electronic preprints of scientific papers on which the proof was published, arXiv, has seen credible research of consistent quality in mathematics, physics, computer science, and quantitative biology. arXiv has even replaced traditional journals as the primary means of communication between researchers in some of these disciplines.

Much of the information technology revolution has also been driven by the same paradigm of open contribution. About 70 percent of all web sites are powered by the Apache web server, free software developed and supported by an open community of programmers. Anyone has access to its source code and can suggest modifications, and it is then up to the team of developers at the Apache Software Foundation to decide whether these changes merit inclusion in official software releases. Sometimes people try to submit “junk code” and vandalize Apache, but such submissions rarely pass preliminary stages of review and certainly never affect any final product.

Despite concerns about quality control, open-contribution projects such as arXiv, Apache, Linux, and Wikipedia have all competed well against similar proprietary initiatives for this reason: Established or presumed credibility is the main metric by which members of open cooperatives decide how much weight to attach to any contribution. Perelman’s bold claim of having proved Poincaré was taken seriously at its outset only because Perelman was already an established mathematician. Such attention would not be accorded to a novice, whose work would be met with skepticism. This hierarchy is a huge incentive for serious scientists to submit only preprints of the highest caliber and for community denizens not to ruin their reputes by publishing false or misleading information.

Each of these projects also has an efficient administrative system with which to stop attempts at sabotage, and it is all but impossible for any mischief to cause harm before it is spotted and thwarted. Any changes or contributions made by a vandal are quickly marked, reverted, and deleted, and may even result in the perpetrator’s expulsion from the community. PLoS ONE will have all of these self-regulating mechanisms, plus an additional financial disincentive for scientific delinquents: Authors have to pay to publish.

We do not claim that online communities should replace more institutionalized peer-review processes in a comprehensive manner. Prestigious journals such as Science or Nature give groundbreaking scientific results the esteem they deserve. But by and large, anxieties about open-publication online journals are unfounded. Even Nature allowed the public to judge papers submitted last summer in tandem with traditional peer-review. Initiatives such as PLoS ONE will help promote free and unfettered scientific study, supplementing and revolutionizing an oligarchic academic process. It is both ignorant and regressive to reject this democratization.



Yifei Chen ’09, a Crimson editorial editor, is an economics concentrator in Cabot House. Patrick Jean Baptiste ’09, a Crimson editorial editor, is a biochemical sciences concentrator in Cabot House.