Academic Poll: Talk Prep

I’m giving a talk at the AAAS meeting next month on international physics tests, and they have asked me to provide information that they will duplicate and distribute to the media. Items requested include:

— A one-paragraph biographical sketch (not a C.V.)

— A lay-language summary of your talk, beyond the abstract.

— The full text of your talk or a related (ideally recent) technical
paper, either as a Word file or a PDF. PowerPoint presentations are
acceptable, but a full text will better serve reporters’ needs.

The first two are no big deal, but the third is kind of weird. I don’t usually make my talks up a month in advance, and I never write out the full text. If I’m doing something complicated and new, I’ll do a detailed outline first, then make slides to fit, but most of the time, I just have an idea of what I’m going to do, and the slide-making an outlining are sort of simultaneous processes.

This is partly my own personal style– I write in sort of the same way– and partly a result of training. My Ph.D. work was done back in the days of overhead transparencies, and my advisor used to write out his slides for a talk on the plane to the meeting. The culture of my part of academic physics holds this sort of thing as the norm.

Which makes this a decent excuse for as poll:

And with that, I’m off to generate some PowerPoints to send to AAAS. Play nice while I’m doing that.

9 thoughts on “Academic Poll: Talk Prep

  1. With my vote, all nine respondents have said that they make an outline and slides, and that’s it. I can imagine that there are some scientists out there who write the text of their talk verbatim and just memorize and regurgitate, but I wouldn’t want to be in their audience.

  2. I don’t make slides. As a being of pure intellect, aka mathematician, I refuse to give talks unless there are at least two large whiteboards (gold standards is of course 9 large blackboards in a 3×3 grid which you can move up and down).
    Slides are reserved for people with large formulas in their talks; 2-commutative diagrams (see some very simple ones at are way easier to draw with your hand.

  3. I can sort of see why somebody who plans to give a talk in a language in which he is not fluent might want to write the talk out ahead of time. Or if you are one of the unfortunate souls who persist in science despite the lack of any extemporaneous speaking ability. But for most scientists, writing out the talk is more trouble than it’s worth. The main reason is that when your talk turns out to be too long, it’s harder to edit a complete written-in-advance talk than to adjust your extemporaneous speech to fit the time allotted. Perhaps the reporters have become too accustomed to political speeches, but if you are a politician of any importance, you aren’t writing speeches yourself (as most scientists would be expected to do), but rather paying somebody to write them for you, and that somebody will know how to write the speech so that you cover the essential points in the allotted time.

    I also came of scientific age in the overhead transparency era. I learned to outline my talks simply because viewgraphs were a scarce and expensive resource, so I didn’t want to waste a slide on a bad figure or a figure I wouldn’t have time to show. Thus I knew roughly what the content of each slide would be before I made the first slide, and I was sure that I wouldn’t have too many slides. In the PowerPoint era I haven’t been as rigorous about that–it’s one of the bad habits that PowerPoint encourages, and one of the more common results is people putting too many slides in their talks.

  4. Like Bashir, I just make slides. I don’t outline anything first. Sometimes sparse slides can be “outlines”, but usually I just make them all in order as they come to me, inserting or rearranging a few later. I don’t write out, memorize, or otherwise plan any text. I just explain the currently displayed slide as it comes up.

  5. When I give a talk,I first do the slides but I also do a dry run or two to get the timing right – nothing worse than slides up for too long or to short a time.

    I have started to use the awesome speech to text programs to transcribe these sessions. If someone wants the (almost)full text of the talk, it is necessary to edit the file before release,
    of course.

  6. I bang together some pictures a few hours in advance, and then say whatever springs to mind.

    Not surprisingly I’m not longer in science …

  7. I just make the slides. Occasionally I have brief comments for myself accompanying a slide, but I never write an outline of a talk or anything like that. As you said, thinking about the story and making slides are simultaneous.

Comments are closed.