There seems to be a small industry these days, consisting of telling us that scientists either or both a) are bad communicators and b) should be doing more communicating with the general public. The latest round is spurred by the articles from Chris Mooney and Sheril Kirshenbaum appearing in many venues as part of promotion for their book Unscientific America. (You can track some of it, in between the feud with PZ Myers, at their blog The Intersection and the articles they reference.)
I confess some amusement at being told both that I (as a scientist) am both bad at something, and should be doing it more. Unless it's something for my own entertainment (singing in the car, for instance), or health (running), I generally avoid doing things that I'm bad at. And think that most people are similar in that approach. But that's an aside to the more important matter.
The continual failing, including from the current round of Mooney and Kirshenbaum articles, is that they (and others making similar comments) never do get around to telling us (scientists) just what it is we should do differently, nor where we should be doing it. Fine-sounding words are mentioned, like 'reach out more'. But, let's try to carry that out. Er. What does it mean after all? Reach out more sounds nice, but after reading The Intersection for quite some time, and asking this question routinely, I still don't know what they mean. I do know that they've written routinely that scientist blogs are not the answer, so my efforts here are a waste by their lights. That's fine, their opinion. (Though I still wonder what, exactly, the question is.)
So I'll turn the question to my gentle, and not-so gentle, readers and ask you for what _you_ would like to see scientists (me as an example) do more of. And, what you'd like to see less of. More blog posts, fewer but better-crafted, more scientists writing blogs, drop the blogs and write letters to the editor, buy TV time and make science-o-mercials, ...?
In like vein, what are some of the dos and don'ts about communication you'd suggest? I do make some effort, for instance, to use a more common vocabulary and avoid math. But ... I do still use math, and know (following the quote from Stephen Hawking, who was advised when writing Brief History of Time that each equation would halve his sales) that this isn't the thing to do for widest readership. Even better if you can point to examples of good, and bad, ideas that I've carried out (intentionally or not) in the blog here.