Matthew Might has an excellent article on his website.
But what are some other good suggestions? One important thing is identification (some professors are simply happier to reply to emails than others). Surprisingly enough, I find that famous scientists in my field (for whatever reason) tend to be more responsive to emails than non-famous scientists. I don't know why this is the case, but maybe it could be that they simply know more, so it doesn't take them as much effort to give information/advice?
Sometimes, I even go as far as to look at Rate my Professors site ratings, since the ones with exceptional ratings often tend to be happy replying to emails.
Another thing: maybe taking a class with the professor (in which case they may feel a bit more obligated to reply to them?) Or maybe simply distributing one's emails in such a way that one wouldn't have to email each professor more than an interval of once every few weeks?
All of the points in the linked article are very true in academia as well. Some points which may be specific to academia which he didn't mention:
- If you're referencing a specific article or talk, reference it very specifically ("In your presentation at SfN 2012 entitled 'The Amygdala and You', ..."). If you can quote a specific slide/paragraph, all the better.
- Mention (concisely, one sentence or less) why you're interested in knowing. Your problem may very well be something which the researcher is interested in, and s/he may be more interested in responding and following up later on.
- If you're working for someone in the field, mention it at the beginning ("I'm a postdoc working for , and...")
- Note that researchers are more often than not interested in talking about their work, so the very first point from the article ("Don't email") is likely bad advice here. Definitely try to get in touch and talk about your mutual interests (if you have a valid question, of course).