An academic term used, essentially, to refer to how people talk on the
Internet.
Studies in computer-mediated communication were being conducted as early as the mid-1980s. These studies were almost universally wholly linguistic in nature; after all, the Internet wasn't that much more than IRC, bulletin boards and email at the time anyway. Even at this early stage, though, it was clear that CMC would prove to be a rich field of study. When you write an e-mail, are you using the register of writing or speech? What about when you're using IRC or AIM? For that matter, what about when you node? The answer, of course, is 'neither' for all three; which register each one is closer to is the subject of much early research into CMC.
As the World Wide Web and more interactive technologies developed, though, the focus of CMC broadened. Aspects of psychology, sociology, and network theory came into play, as academics started analyzing the languages of text messaging, Facebook, and VoIP. At the same time, as a solid foundation of research was established, more particular subfields--most notably, gender studies--were investigated. Today, a quick glance through the Journal of Computer-Mediated Communication (published online and freely available at http://jcmc.indiana.edu/) turns up studies on topics ranging from "Google Bombing From A Time Perspective" to "Islam, Jihad, and Terrorism in Post 9/11 Arabic Discussion Boards" to "Greetings and Closings in Workplace Email".
Despite the interdisciplinary nature of CMC, linguistic research remains at its core. Here's a list of nodes describing research into various subfields of computer-mediated communication, with a strong linguistic emphasis:
And some current CMC scholars, though nodes on these may be a while: