The extended implications.
All of science is built by standing on the shoulders of those who came before. Scientists traditionally give their work to the world for free, but this isn’t purely altruistic. In part, it’s a pay-it-forward deal; giving your work away is the price you agree to pay in return for the free usage of the science of the past.
Closed-access for-profit journals fundamentally violate this ethos. Even if they were run ethically, they would still be objectionable.
But they aren’t run ethically. Corruption scandals amongst the for-profit journals are routine and increasing; see http://www.badscience.net/2009/05/elsevier-get-into-fanzines/ for an example. In the medical sciences especially, this corruption has literally murderous consequences.
Beyond that, for-profit journals create many bad externalities. Academic institutions in the developing world cannot afford them; this in turn cripples the ability of local researchers to respond to the medical, environmental and social problems of which their communities have the most intimate knowledge. The resulting suppression of academia and brain-drain of local scientists towards the developed world further damages their societies.
For-profit journals also increase the tendency towards sensationalism, the neglect of replication and the bias against publishing negative results.
Although there is some cost involved in journal production, almost all of the work is done at the expense of unpaid junior academics. Not just the writing and reviewing; these days, academics have to do the bulk of the typesetting themselves as well. Journal submission processes are nightmarish; getting a manuscript into an “acceptable” format takes days of work, and uploading a submission often requires hours.
It is entirely within the technological ability of 21st century humanity to create a single database, paired with an excellent search engine, that includes the full text of every single academic paper that has ever been published, and to make this database freely available to every person in the world. If the cost were shared amongst academic institutions worldwide, the expense would be trivial.
Doing this would constitute the greatest advance in the dissemination of human knowledge since the Gutenberg press. The only reason why we can’t do it is because a bunch of hugely profitable rent-extracting parasites are squatting on top of an unearned mountain of copyrights.
