Lance Fortnow has an interesting post on why conferences and conference proceedings are so much more important in CS than in older fields like physics.
Lance’s post started me thinking about a topic I wonder about from time to time: the role preprints play in science. At the moment, the main role journals seem to play in physics is resume-building: for grants, hiring and promotion it’s important to get into the so-called “best” journals, like Science, Nature, and Physical Review Letters. Peer review is often mentioned as another function of journals, but there are other ways that can be accomplished (see, e.g. Daniel Gottesman’s sadly defunct preprint review site.) Resume building does not seem to me to be a sufficient justification for the incredible amount of money spent on journals.
One of the (frankly unconvincing) arguments that journals are worth the money is their editorial role. This is related but distinct from their role as a stamp of approval for authors, which you mention, since it is a service to readers rather than authors.
It aims to eliminate stuff that is either wrong or not interesting. There is no doubt that this costs money, but that it is of any use to the modern scientist if less obvious. Certainly the explosion in published material and increased specialisation makes it impossible for any scientist to wade through it all themselves and assess it for interest and correctness. This is nominally where the “best” journals come in. However, in physics at least, I get the impression you can keep up-to-date by keeping an eye on arXiv, and via personal recommendations.
The question of whether journals are worth the money to scientists is different from whether the prices are a fair reflection of the cost. Although journals do cost an incredible amount, so does running a publishing businnes. Especially when your publications have a circulation of order 10^2 and a staff of of order 10^1!
In the particular case of society publishers, where the excess goes back into the subject this amount is less difficult to justify. E.g. Physical Review’s profits go into physics in the US via the APS.
Conveniently, the UK’s House of Commons Science and Technology Committee is due to publish its report on Open Access in just a few hours, and will doubtless have something to say about value for money. The report will appear at http://www.publications.parliament.uk/pa/cm/cmsctech.htm at 0001 UTC.