Powerpoint

Doron Zeilberger gives the best reasons I’ve seen for eschewing Powerpoint in favour of whiteboard talks. His reasons are pretty pertinent for detailed technical seminars where you want to understand ideas in detail. I still think Powerpoint has a place when you’re just trying to communicate the gist of your results, as is often the case in conference talks, and a higher baud rate is appropriate.

6 comments

  1. I wonder. It seems to me that communication ability is independent of the technology used, overhead, whiteboard, PPT, or even the old-school chalkboard. After all, if you can’t present decently, your presentation will go straight to dev/null no matter what technology you used. Just think about the worst lecturers you had in school and whether PPT would have made them better or worse; probably neither, you’d most likely just remember that they were horrible.

  2. You may find Edward Tufte’s The Cognitive Style of Powerpoint interesting. It’s $7, but worth every penny. If I understand your post correctly, you’re saying the rate of information transfer of a Powerpoint presentation is higher than that of a traditional talk using the blackboard or handouts. Tufte’s thesis is the exact opposite and his evidence back this up. It’s best to read it yourself (it’s a vigorous 27 pages), but amongst his assertions are that typical Powerpoint presentations are extremely low on data and that the various display methods (particularly bullet points) break the semantics of what you’re saying. I can’t recommend it highly enough.

    http://www.edwardtufte.com/tufte/books_pp

    He was cited by the Columbia accident Investigation team in their report into the reasons for the disaster. Apparently, the lack of communication caused by Powerpoint presentations lead directly to a failure to take positive action to prevent the accident. (Lot’s on this here: http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0000Rs&topic_id=1&topic=Ask%20E%2eT%2e — these are two pages taken from the essay itself).

    There’s also the famous (and funny!) Powerpoint Gettysburg Address http://www.norvig.com/Gettysburg/ and Aaron Swartz’ Powerpoint version of … The Cognitive Style of Powerpoint http://www.aaronsw.com/weblog/000931. Googling for ‘Powerpoint makes you dumb’ will throw up lots more on this.

  3. Mike Williams: “If I understand your post correctly, you’re saying the rate of information transfer of a Powerpoint presentation is higher than that of a traditional talk using the blackboard or handouts.”

    I guess what I’m saying is that the rate at which information is _presented_ (not transferred) is higher during many PPT presentations of technical seminars, when compared with blackboard talks. Based on what I’ve seen of Tufte, and I’ve only glanced through a couple of his books, I think he’s primarily concerned with much less technically oriented talks than Zeilberger and I are in our posts. I’ve seen some amazingly complicated mathematical proofs “explained” in an hour with PPT, the sort of stuff that might have taken 5 or 6 hours to cover at a blackboard, with an audience of experts. Too bad nobody understands anything after the first 10 minutes of the PPT presentation…

  4. “I wonder. It seems to me that communication ability is independent of the technology used, overhead, whiteboard, PPT, or even the old-school chalkboard. After all, if you can’t present decently, your presentation will go straight to dev/null no matter what technology you used. Just think about the worst lecturers you had in school and whether PPT would have made them better or worse; probably neither, you’d most likely just remember that they were horrible.”

    I agree, but this does not mean that the converse is true! I.e., a person who potentially may give a good talk may screw it up in the wrong media.

    Zeilberger’s post struck a chord with me because of my recent experience in running a seminar series. After some experimentation, we’ve done three main things (and a half dozen smaller things):

    (a) Banned audiovisual aids, except the whiteboard;

    (c) Advised speakers to prepare 40 minutes of material for a whiteboard talk, for a 75 minute slot; and

    (b) Occasionally, a prize will be given (wine and beer are popular choices) to the person who asks the most questions.

    After six months of the seminar series, a typical week has about 30 questions asked, and I think a large fraction of people go away understanding most of what was said in the seminar, right down to highly technical details. Asking people to give whiteboard talks is only a part of this, but I’m pretty sure it’s a very important part, as it slows things down enough that people have a chance to follow, and ask sensible questions.

  5. Hi Michael,

    I think you’re right that Tufte’s ideas are on much shakier ground when applied to maths and mathematical physics. In such contexts Powerpoint is often used as little more than an OHP, and there’s certainly a lot to be said for it (and other software such as Apple’s Keynote and OpenOffice) when used as nothing more than a portable and convenient alternative to transparencies.

    It’s when I see extremely complex ideas decomposed into bullet points, with the simplification of relationships and context that necessarily entails, that I start to worry (or just switch off!). For extremely disorganised speakers this may be useful in clarifying the structure of their own argument. That’s fine for them because they are already familiar with the complexities of their talk. The details are implicit. To the audience however, the list is usually at best an abridged version of the argument, and at worst positively misleading. I can’t help feeling being short changed by such talks. YMMV 😉

    By the way, I’m really enjoying the weblog. It’s very stimulating reading for a lapsed physicist like myself.

  6. Whilst on the subject of how to run a conference, I like Joel Lebowitz’s method at his annual statistical mechanics conference at Rutgers. New students in particular are encouraged to give extremely short introductions to their work. Some time ago they were allowed as much as ten minutes, but the conference is now so popular that they are restricted to much less (three minutes, I think). This limit is ruthlessly but politely enforced by Professor Lebowitz himself.

Comments are closed.