From the articles linked from Welcome to Less Wrong:

1) http://lesswrong.com/lw/jx/we_change_our_minds_less_often_than_we_think/

The title is descriptive and the text is short and to the point. Empirical support is present and clearly stated. Of course it could be shortened quite a bit more without losing any information, but I don't find it excessively verbose.

2) http://lesswrong.com/lw/qk/that_alien_message

Its a long post, not trivial to follow, and when reading its not clear how the effort will pay off. Perhaps this is evidence of a short attention span, but I've generally found that most concepts can be expressed succinctly. It might also be a habit of my profession that I try and make writings as terse and general as possible.

I suspect status and article length are highly correlated (e.g., people read autobiographies of famous people), and so longer writings might be ways to signal status.

I can produce more examples, but the above two are archetypal for me.

3) Well, I don't know what I don't know ;) But to list a few things:

• Pros and cons of frequentist vs. bayesian approaches. Everything I read here seems pro-bayesian, but other (statistics) sites I look at promote a mix of the two approaches.
• Why so little discussion of mechanisms which improve the rationality of group action and decision-making? Is that topic too close to the mind-killer, or have I missed those articles?
• I find appeals to rationality during strictly normative argument irrational, because people don't seem to adopt ethics on the basis of rationality or consistency. Thus I'm confused by the frequency of ethical discussions here. Am I missing something about ethics and rationality? Or just wrong? Something on a general rationalist approach to ethics would be helpful to me.

Agreed. I'd be quite interested which post non-LW friends found useful (and if they passed them on to anyone else). My mom ended up using Twelve Virtues as a discussion reading in the first day of her eled class (elementary education).