Back in December, a short bill was introduced in the House called the "Research Works Act". Its backers, Darrell Issa (R-CA) and Carolyn Maloney (D-NY), describe it as something that will maintain the US's standing in scientific publishing. After looking over its language and reading a number of commentaries on it, I have to disagree: this looks to me like shameless rent-seeking by the commercial scientific publishers.
And it pains me to say that, because I know several people in that business. But it's a business whose long-term model has problems. (See the Addendum below if you're not in the field and want a brief summary of how scientific publishing works). The problem is, the work of the editorial staff has changed a good deal over the years. Back when everyone sent in hard copies of papers, in who knows what sort of format, there was a good deal of work to do just turning the good ones into a consistent journal. Electronic submission has ironed a lot of the grunt work out - it's still work, but it's not what it used to be.
That leaves the higher editorial functions themselves, and here's where the arguing starts. Most, and in some cases all editing of content is done by unpaid peer reviewers. There are journals whose editors exist mainly to keep the flow of submissions moving to the reviewers, and from them back into the official journal, while hardly ever laying a finger on the copy itself. They function as Peer Review Mailroom Managers. And while that's a necessary job, it's the center of the argument about scientific publishing today. How much, exactly, is it worth?
Scientific journal are expensive. I mean, really, really expensive to subscribe to. And if you're not a subscriber, access to individual papers is pretty steep, too - typically in the $15 to $50 range. This is the business model for commercial scientific publishing: create a space with value (reputation, name recognition) and charge the maximum that that traffic will bear. And that's fine; there are a lot of businesses that work the same way - if they can.
The problem is, the information-sharing capabilities of the Internet blow a large hole in some of the traditional publishing model. And another problem is that a large number of papers that come into the journals from US academic researchers have had some (or all) of that work paid for by government grants (NIH, NSF, DOE and so on). As it stands, articles funded by the NIH are available in PubMed Central for free access, no later (by law) than 12 months from the initial journal publication. Researchers can also submit their work to "open access" journals (such as those from the Public Library of Science), which charge a fee to authors to defray editorial costs, but then allow immediate unlimited access to all comers once a paper is accepted. (I should note that some commercial journals get away with "page charges" as well, and some have a model where the authors can pay extra to bring their paper out from behind the paywall).
And here's where we have the Research Works Act. It would forbid any publication in an open access journal for anything funded in academia by US government grants, and it would forbid any public-access repository for such work. That's its purpose. Well, to be more accurate, its purpose, as described by the head of the Association of American Publishers, is that it "ensures the sustainability of the industry". Yep, make my business model part of statutory law, and beggar my competition: what else is a government for, anyway?
Update: see the comments section. I'm interpreting the text of the law to mean the above, but another way to read it - probably the correct one - is that it's mainly rolling back the 2008 law that mandates that NIH-funded papers go open-access after a year. But that's bad enough as it stands.
To their credit, the MIT Press looks like the first big academic publisher to defect from this position. But the commercial publishers (Elsevier, Wiley, and so on) will never give up on this goal. Yes, the RWA, according to them, is aimed at "preventing regulatory interference with private-sector research publishers". Here's Congresswoman Maloney using Elsevier's own press release language, sentence by sentence, as detailed by Michael Eisen, co-founder of PLoS. (He also has an op-ed in the New York Times on this issue).
I see no reason why we should make the current scientific publishing system a matter of law. I think it should change - and be allowed to change - as new technology allows it to. And I think that the Research Works Act is nothing more a blatant attempt to hold on to a profitable business plan.
Addendum: For those outside the scientific world, here's a brief summary of how things have traditionally worked. As a scientist (academic or industrial), you take the time and effort to write up your results for a journal. You have to pick your journal at the start of the process, since each of them have their own ways of organizing a paper, their own preferred way of citing other papers as references, and so on. Anyone can send anything to any journal they feel like, although you'd be well advised to target your paper to the ones that (a) have the best chance to actually accept it and (b) will do you good to have a paper published in. The overlap between those two may not be large, or may not exist at all, depending on your paper. These days, most journals have templates for Word or the like, which standardizes the submissions, and have some sort of automatic PDF generation during the submission step step so you can see how the paper will look when formatted in the journal's style and page layout.
Your manuscript is given a quick check to make sure that it's appropriate for review. A few of of the higher-end journals make this a key step, because they can afford to turn down even rather interesting papers as not necessarily worth their time to go on checking. But in most journals, unless there's something obviously off, your paper goes out for peer review. Two or more scientists from Out There Somewhere look it over and send in comments. Those comment forms have a section for the original authors to see, and a section for remarks that go just to the editorial staff, and you can use those as you see fit. (I, for example, once used the latter forum to ask the editors to please stop sending me papers from a certain author, because I'd done three of them and couldn't stand to see any more. They honored my request.)
As an author, you see the comments when they come back and get a recommendation from the journal - usually it's "Publish as is", "Publish after minor revisions", "Publish after major revisions", or "Go away". That last one usually isn't expressed in quite those words. The middle two are the most common, since most stuff eventually gets published somewhere if the authors are persistent enough (and are willing enough to have their work appear in the Zambodian Journal of Chemistry or what have you).
Now at this point, traditionally, the work of assembling the accepted papers into a printed journal kicks in on the editorial side. And it still does, but that process is becoming less and less important. I honestly can't tell you when I last saw a hard copy of any of the journals I read regularly. Even the idea of separate issues is becoming antiquated, since new papers (in the case of many journals) just plop out onto the web site (and into the RSS feeds) as they emerge from the review process.