Getting into Harvard is hard, very hard. Yearly the gatekeepers in Byerly Hall vet thousands of applicants on their merits, rejecting many times the number of students that they accept. But getting a scientific paper published in Science or Nature, today’s pre-eminent scientific journals, is oftentimes harder.

Science, like much of academia, has its own admissions committee. Though over a million manuscripts are published in journals yearly, many more are submitted and rejected. The gatekeepers of science—peer reviewers who are reputable scientists and well versed in a particular field—advise journal editors whether to reject a manuscript outright, send it back for revisions, or publish it. And publication is everything in science. If an experiment doesn’t appear in print, it might as well have never been performed. But the peer review process, even to researchers, can seem like a black box. Since the scientists who review the submitted papers review them anonymously, there is little accountability—these gatekeepers, some think, have far too much power over the progress of science.

For these reasons, some in the scientific community have proposed switching to open-publication, online journals. In one model, used by the soon-to-launch online journal of the non-profit Public Library of Science (PLoS), scientists will be able to publish their papers online for a fee after only nominal editing by the journal’s editors. The review process would take place online and post-publication. Only in this case, anyone, not just the author’s scientific peers, would be able to post comments and reviews.

In theory, such a system would transform the scientific community into true “marketplace of ideas,” in which scientific results would be vetted democratically, instead of by a group of cloistered elites. Readers would post comments judging the quality of the work, and an experiment or theory would either be buoyed by praise and interest, or, if found flawed, drown in a sea of anonymity. Such a system, proponents argue, would free science from the trammels of communication that currently retard its process. Chris Surridge, PLoS ONE’s managing editor recently told the Associated Press, “If we publish a vast number of papers, some of which are mediocre and some of which are stellar, Nobel Prize-winning work—I will be happy.”

And therein lies the problem.

Without a peer review process to separate the revolutionary papers from the merely good from the rubbish, scientists will have no way of knowing which discoveries and experiments merit their time and interest. Instead, they will spend inordinate amounts of time wading through the quicksand of junk science to get to truly interesting work. Peer reviewers are chosen as peer reviewers for a reason—unlike the hoi polloi that roam the Internet, they have the knowledge and experience to judge scientific research on its merits. Furthermore, the peer review process strengthens papers, as authors are forced to defend their research’s weaknesses—without peer review the temptation to prematurely publish incomplete or uncontrolled experiments will be formidable. Far from “a marketplace of ideas,” open publication will create a morass from which science might not emerge. Results will be duplicated, communication retarded, and progress slowed to a standstill.

Proponents of open publication who point to advances such as Grigori Perelman’s recent proof of the Poincaré conjecture, which was posted online instead of submitted to a journal, fail to realize that such instances are the exception rather than the rule. True, the traditional peer review process is not perfect. It delays the flow of information, can sometimes be biased, and often unduly prioritizes the work of established, famous scientists over the work of lesser-known researchers. But it is far better than the alternative.