This debate has been going on for years. Those who decry the cost of publication and try to evade it have all discovered the same thing: even with volunteer reviewers, vetting, formatting, and maintaining papers securely online is expensive. Most of the Public Library of Science (PLoS) journals, for example, now charge authors $2250-2900 per article. Oxford University Press generally charges even more, and still claims to be losing money. Often, authors are hit with additional "excess" page charges beyond that fee, because their paper has expanded due to additional data demanded by reviewers. If a typical 5-year $200,000/year grant results in 12 papers, that means 3-4% of the funds are devoted just to publishing the papers. As a scientist myself, I was initially excited about the open-access idea, but I'm no longer convinced that it's any better or less expensive than the old system of private publishers. It just means the costs have been shifted from subscribers (mostly university libraries) to scientists and their laboratories, who in general can ill-afford it. "Taxpayers" already have access to all articles through PubMed within a year after publication, and they have access to the abstract (summary) immediately, which is usually as much information as they can use. Still, you wonder who is going to pay $30 just to look at one article and whether the journals wouldn't make more money if they charged $2, or something low enough so that buying it might actually be worthwhile.