Why not? Because I rarely like to break any sort of rule:
Guy Leonard received an unpleasant surprise in his inbox early this morning: a notice from Academia.edu saying it had taken down a copy of an article of his that he’d posted on the research-sharing platform. The reason? A takedown request from Elsevier, which publishes the journal in which the paper had appeared.
Mr. Leonard, a research fellow in the University of Exeter’s College of Life and Environmental Sciences, tweeted his dismay and posted a link to a screengrab of the notice. ...
Richard Price, the founder and chief executive officer of Academia.edu, said in an email that “Elsevier has started to send academics on Academia.edu takedown notices in batches of a thousand at a time.” The email Mr. Leonard received “is the notification that we sent to our users,” Mr. Price said, adding that his company usually receives one or two individual notices from publishers a week, “but not at scale like this.” (Academia.edu has close to six million registered users; it said it had received about 2,800 takedown notices from Elsevier so far.)
Mr. Leonard was not the only researcher to receive such a notice this week, as Michael P. Taylor, a paleontologist and open-access advocate, reported in a post on his group blog. Many researchers post copies of their articles online, Mr. Taylor said, even if they’re not legally supposed to. “It’s always been so, because even though technically it’s in breach of the copyright transfer agreements that we blithely sign, everyone knows it’s right and proper,” he wrote. “Preventing people from making their own work available would be insane, and the publisher that did it would be committing a PR gaffe of huge proportions.”
via chronicle.com
I prefer an open-access world almost as much as anyone, but it seems like an author should respect the forms that they sign. Most of these forms allow you to share your PDF with interested colleagues via email.
Full disclosure: I'm in the last month [!] of service as an associate editor at an Elsevier journal.