I decided to sign up for twitter because of the benefit of being able to tweet thoughts and questions about the research during the sessions. Twitter was... inconsistent. I'm not sure that I will use twitter much outside of meetings, but I think it is a fantastic, and potentially very powerful tool to use during scientific meetings.
Hashtags can be used to collect the thoughts, comments and questions from the audience, and save them to be reviewed later. Discussion is facilitated, and potential collaborations introduced by shared interest that arises from comments during the seminars. Oftentimes there is only time for one or two questions after a presentation, but surely many, many more questions remain unanswered in the audience. Twitter, potentially, is a venue for some of these questions to be brought up and addressed.
Maybe Twitter isn't the best venue, because of its character restriction, and also because of its inconsistency, but it is a good start for now. The problem that I encountered with Twitter is that the "searchability" of my hashtags was very inconsistent. The first posts I tagged showed up in the feed, then the afternoon posts did not work. The next day again my hashtags showed up for only a few sessions, and then did not show up. I am still trying to figure out what the problem with the hashtags is, but in order for Twitter to be the useful tool it could be for conferences, it needs to be figured out.
The benefit of a character limit is that it ensures attendees won't be so easily distracted during the seminars; one can simply tweet as the thought arises, similar to taking notes (which I do on my laptop anyway). Only now these thoughts can be shared with the rest of the community.
If you'd like to see the comments from the meeting, search twitter for #bg2010. Some of my posts are there. :)