I had been looking forward to watching the EY’s presentation, and was bitterly disappointed.
On stage he looks like a novice speaker, committing a slew of rookie mistakes: most of the jokes fall flat, the slides are uninformative for anyone who is not already an expert in the area (and useless for anyone who is), he constantly complains about time constraints (did they shorten his time unexpectedly at the last moment?), he does not leave time for a conclusion, he talks at a level likely inappropriate for the audience, he flashes a score of slides toward the end to show that he’s done more work than he was able to present…
Now contrast this with the presentation that immediately follows, by a seasoned pro, Max Tegmark, who knows how to be engaging and hold the audience’s attention and how to get his point across, whose slides are well thought out and whose jokes elicit the intended reaction.
I am loath to give advice to a person who is orders of magnitude smarter than I am, but, were it anyone else, I would recommend a basic presentation-giving training, such as the one I took ages ago.
The talks for SS06 and SS07 were presented much better, so it seems like a matter of investing less effort in preparing for the talk (perhaps with less-time-than-expected making the challenge of happening to be prepared anyway harder), rather than of not being able to present well.
Which is why I explicitly went out of my way to modify the saying to ‘this’, sacrificing catchiness so it was clear to readers that I was not asserting “If a job is worth doing, it’s worth doing well” as a general claim. Without success, it would seem.
It seems to me that Eliezer’s goal was to use the talk as a lure for any potential FAI researchers who might be in the audience, and simultaneously to signal that SIAI is doing real technical work and not focused purely on PR or “spend all our time pondering morality”. For these purposes, a really smooth practiced presentation would not be particularly helpful.
I had been looking forward to watching the EY’s presentation, and was bitterly disappointed.
On stage he looks like a novice speaker, committing a slew of rookie mistakes: most of the jokes fall flat, the slides are uninformative for anyone who is not already an expert in the area (and useless for anyone who is), he constantly complains about time constraints (did they shorten his time unexpectedly at the last moment?), he does not leave time for a conclusion, he talks at a level likely inappropriate for the audience, he flashes a score of slides toward the end to show that he’s done more work than he was able to present…
Now contrast this with the presentation that immediately follows, by a seasoned pro, Max Tegmark, who knows how to be engaging and hold the audience’s attention and how to get his point across, whose slides are well thought out and whose jokes elicit the intended reaction.
I am loath to give advice to a person who is orders of magnitude smarter than I am, but, were it anyone else, I would recommend a basic presentation-giving training, such as the one I took ages ago.
EDIT: That course is also where I first learned how to structure oral and written presentations.
The talks for SS06 and SS07 were presented much better, so it seems like a matter of investing less effort in preparing for the talk (perhaps with less-time-than-expected making the challenge of happening to be prepared anyway harder), rather than of not being able to present well.
That being the case it may have been a time to consider the heuristic “if this is worth doing it is worth doing well” and apply modus tollens.
It’s a catchy saying. It may or may not be a good heuristic.
It certainly doesn’t hold in every case.
It might have in this case.
Which is why I explicitly went out of my way to modify the saying to ‘this’, sacrificing catchiness so it was clear to readers that I was not asserting “If a job is worth doing, it’s worth doing well” as a general claim. Without success, it would seem.
It seems to me that Eliezer’s goal was to use the talk as a lure for any potential FAI researchers who might be in the audience, and simultaneously to signal that SIAI is doing real technical work and not focused purely on PR or “spend all our time pondering morality”. For these purposes, a really smooth practiced presentation would not be particularly helpful.
Bitterly? Just what were you expecting?
Having never met him or seen him on stage, I was expecting to have my mind blown, MoR style.