I guess you didn’t have a chance to read it before your talk.

An audience hates when a presented goes over his or her limit and bites into the other speakers time. Other speakers also tend to pare their own presentations and, presto, the takeaways from the panel discussion suffer.

I spent too much time trying to figure out how to present the One-New-Thing. I settled on presenting it as “One-Big-Joke”.

Turns out that, after sitting through 30 talks, the audience were ready for it. I don’t konw if the audience remembered the One-New-Thing — but they did remember the One-Big-Joke, because one of them still needle me about it a year later.

YMMV

Find out what parts they didn’t understand, and fix them.

Then, after fixing the colleague doesn’t work, fix the parts they didn’t understand.

My 5-minute talk literally only had 4 slides, if memory serves. One was the title page. One was a one-slide description of the method we used for finding supernovae — it had a supernova lightcurve with a few points highlighted on it, and I talked about how we found them and where. One was the Hubble Diagram, showing that, yes, we had magnitudes and redshfits of lots of supernova to plot. The most important one was the preliminary version of the Omega-M Omega-Lambda confidence intervals, which showed all of the confidence floating above the Omega-Lambda=0 line.

The saddest thing about all of that this was January, and we didn’t write a Letter or some such that we got submitted later that month. The actual paper was a magnum opus, and wasn’t submitted until something like November, and has a 1999 publication date on it. The other team saw our preliminary results, lit a fire under themselves, and got their discovery paper out just a few months later.

I don’t think “much bad” can come from a student giving a 5 minute talk. The worst things that can happen – student gives rushed talk, no one remembers student’s performance; or student gets scheduled for Thursday afternoon, nobody shows up. These both suck, but it is not like anyone really gets a black mark on their permanent record for an AAS talk.

I liked going to see AAS posters when there was a decent chance you could actually talk to the person who made the poster. One of my complaints about what the AAS has evolved into is that the poster session room is full of poster-blitzes from large projects, where no people are ever present, so if you are presenting a poster you feel lonely and sheepish for standing around near your poster in an empty room. Apparently some prefer viewing posters in an empty room, but I don’t.

http://www.science20.com/qed_reformulation_feasible/blog/unknown_physics_particle_orbital_momentum

This is usually true for 50-minute talks as well. I suspect the median number of interesting new ideas that are communicated really well to most of the audience is zero, per talk, regardless of how long the talk is.

Few things are worse than trying to cram too much information in with too many slides, and then having to page through them so fast that the audience can’t take any of them in.

I can give a 25 minute talk with 5 slides and not have dead air. So can anyone, if you practice. So I think if you have more than 5 slides for a 5 minute talk, you’re losing the battle. I would prefer fewer.

That would be useful for me… I am also now even more jealous than usual of people who do research involving pretty pictures.

“it should look a lot like a 5 minute talk”

But I disagree with this. People need to be able to understand your poster even if you aren’t there. If people can get everything out of your talk just by looking at the slides, then your talk has way too much text.

But that’s just me. I know some people who love the poster format to present entire ApJ Letters…..