Plus ça change...

publishing in peer-reviewed medical journals has seen dramatic changes in the last 2 decades. First, the introduction of the fax machine and the personal computer changed the way in which manuscripts are prepared, peer reviewed and published. In the early 1980s personal computers at CMAJ were rare, and processing manuscripts was laborious for both authors and editors — endless typing, retyping and proofreading of hard copy as it evolved, step by step, from the authors' submitted version to the final published article. Now manuscripts are accompanied by electronic files, from which articles are edited and typeset on screen and transmitted to the printer by disk or modem. Furthermore, medical information is now disseminated not only through print journals, but also on an ever-growing number of journal Web sites. For all this change, has the quality of information improved? Unfortunately, the electronic revolution has not altered in any fundamental way the most essential component of processing scientific medical information: its validation. Peer review still adheres to the same process that developed in the 17th century, and editing is essentially the same process, be it with quill pen, pencil or electronic keyboard. Only recently have peer review and medical editing become the focus of serious study and research. It was confirmed recently at the third International Congress on Biomedical Peer Review and Global Communication , held in September in Prague, that peer review and editing do, indeed, improve the quality of medical manuscripts. However, both are fraught with imperfections. Peer reviewers and editors are often untrained and frequently miss major flaws in manuscripts under review, and they may have biases that unfairly influence the ultimate publication of manuscripts. And authors often lack the knowledge or the will to present their study findings clearly and honestly. The solution, at least to this observer, lies in concerted education of all medical scientists in the preparation and appraisal of reports of medical studies. It was heartening at the Prague meeting to observe chinks in some of the shibboleths of the peer-review process. For example, editors who have painstakingly tried to ensure that authors and reviewers do not discover each other's identity will be pleased to learn that the task is probably not worth the effort. They will not be pleased, however, by the observation that they are selecting reviewers who do not necessarily have the competen-cies needed to evaluate manuscripts and that they are inconsistent …