Conference organizers must learn to record from the computer more than from the camera. I understand there are some difficulties with different frame formats, but, as a last resort, one could at least record a screencast (QT animation is a lovely format for that) that could be later synced and combined with the audio track.

And it would also free the camera to record the presenter more closely.

It looks like she took the video herself. You can see her set up the camera from her seat before she approaches the stage. The sound is awful, and not having a screencast really reduces the value of this presentation.

I am taking a mental note the next time I do a presentation, to record a screencast at the same time.

I will do some tests with QT animation format and see how much I can do.

Really. They should do a screencast along with a recording of the presenter speaking and they should be presented together. I've seen this down, it can't be that hard.

I gave up after the first 30 seconds of the main talk. Does the audio get any better later on?

I watched maybe eight minutes of it. It's pretty much the same as the beginning -- her voice is in a register that, when encoded, is very frustrating to listen to. Also, she spends much of it apologizing for Emacs's extra features, like Snake, and Psychotherapist. She barely got into the actual examples when I turned it off. It was hard to watch, because for this presentation, I want to see what's on the screen, not the presenter.