Scientific communication is (steadily) changing. Some alternatives to the traditional scholarly communication system are Open Access self-archiving servers like ArXiv, which has been successfully in place for more than twenty years; Open Access publishing, in which authors pay a fee to make their articles freely available to everyone; and Open Notebook Science, in which researchers make their notebooks and raw data public online so that everyone can see/contribute to their progress (for an example of ONS, see this piece of Nature News and this blog).
Despite these changes, in many disciplines, such as chemistry, research is disseminated mainly via traditional scientific journals (see the Nature Chemistry comment “Communicating Chemistry” for possible reasons why chemists are slow to adopt other publication models). Many of those journals are published by commercial publishers and are subscription-based, which means that libraries pay (a lot of) money to have them. This situation leaves publishers in a perceived dominant position.
Recently, the eminent mathematician and Fields medallist Tim Gowers has taken a stand against one of the big publishers of scientific journals: Elsevier (you may read his blog post here). His main objections are: 1) they charge excessively high prices for their journals; 2) they bundle journals, which means that libraries have to buy a large set of journals to get the titles that they want; and 3) they support SOPA, PIPA, and the Research Works Act. This post has stirred the mathematics community, and many mathematicians and other scientists have signed a declaration of unwillingness to support Elsevier’s journals by publishing, refereeing, or doing editorial work. You may read more about the declaration here and find it here.