Publish or Perish

Imagine, if you can, an industry where you are given something that you haven't paid for, you ask other people to edit and comment on it for you for free and then get to sell the result back to the very people who did the work in the first place and who edited it for you.
You might be thinking 'music industry' at this point, and while i'd agree with you, i am in fact describing scientific publishing - the pinnacle of what can be a long and lonely road in getting some original research done. In case the concept of 'getting published' is foreign to you, i'll describe the major steps:

1. Academic applies for funding from large govt science funding body i.e. NIH, EPSRC, or if in NZ, FoRST lets you splash out and buy a new pack of BluTack and some paperclips, not expensive paperclips, approved supplier paperclips...
   
2. Funding is granted and academic buys people/equipment to do the work
3. If you're lucky/determined, some novel work is done that is worth sharing with your peers
4. Write a journal article and send it in to a journal that publishes papers in that field
5. Journal sends your manuscript to anonymous reviewers for comment and recommendation (typically: publish, publish with minor errors corrected, you must be joking, i temporarily lost the will to live)
6. Journal combines successful articles together on a semi-regular basis
7. Publisher of the journal sells the collection to libraries so that people can read them

Pretty cool huh? You get to sit around and decide who gets published and who doesn't and earn several billion dollars a year!
Now we have this inter-webby thing and some academics are starting to ask 'since we subscribe electronically and you have no publishing costs (or at least orders of magnitude lower than they were), why the hell are the subscription fees going up at an astronomical rate?'. Good question.
Bring on the FUD.
The standard response is 'oh yeah?, we do lots of stuff that you don't appreciate, like, forwarding the manuscripts to other academics, and opening their email replies...'. This really is a hornet's nest and to be clear on my position: they are middle-men and deserve to suffer the fate of middle-men through the ages i.e. eventually get cut out of the picture as and when technology allows. Not surprisingly, many not-for-profit publishing houses don't like this nor do their multi-national for-profit brethren.
Check out this from the Royal Society in Britain, yet another reason why i won't join the RSC. This seems reasonable as a third-way, but at the end of the day i prefer the Public Library of Science (where the authors pay for editing but access is completely free).
Can we imagine a day where publishing is as common as a blog entry and we let our peers decide who's good and who's not directly? Getting into a dead-tree edition then becomes something special. Add in some filtering software (i.e. always let me know when A, B or C release something and also whoever they ask to always be told about) and we're good to go. Even with several thousand scientists in a field, it doesn't take that many degrees of seperation before you're covering a significant amount of literature.
This isn't totally academic (no pun intended), China and India will start to rival Europe and the US in technology within a few decades (i doubt before 2050), do you think they're going to want to subscribe to journals at an annual rate that could buy ten research workers? We may find that waiting several months for your manuscript to actually get published is gonna seem antiquated very, very quickly.