Does Powerpoint makes you stupid?
I detest Microsoft Software, and not out of visceral “knee jerk” hatred, but because I used it for a long time and always recognized it as unsophisticated and barely usable. The only way to make it usable was to make things really ugly the way the developers intended, i.e. so that Microsoft made money by creating addicts, instead of helping me to actually create a good piece of work. That’s why I was eager to read an essay by Edward Tufte about the “Cognitive Style of Powerpoint.”
Tufte lays out a compelling case for Powerpoint being not just terrible software, but that Powerpoint has had a dramatic negative effect on the presentation of scientific and technical data. While I agree with much of what he says, I was disappointed that I actually disagreed with him about a lot. Disagreement in itself is a good thing, but I genuinely expected him to offer data I could use in my crusade against Microsoft Office. Instead he offered some disappointing and impractical alternatives, based on a weak argument and sometimes clever anecdotes.
1 Powerpoint is bad
I agree with Tufte’s general sentiment that Powerpoint is bad, and it’s bad on multiple accounts. First of all it’s proprietary software, and therefore nobody should use it in the first place. Beyond that (there’s more!) Powerpoint just sucks. Anybody who’s used it for more than five minutes is experiencing what’s wrong with it. It’s poorly designed under the hood and on the dashboard; it mixes structure and content specification, and does not encourage users to make the distinction between structure and content. The output of it just looks like crap, too: low resolution, ugly fonts (like Arial) and poorly proportioned, obviously unprofessional typesetting (if you can even call it “typesetting”). Steve Jobs was right that Microsoft has no taste; I agree with both him and Tufte on this point.
Furthermore, Powerpoint should not be used as a substitute for a good paper or technical report. Tufte makes this case quite compellingly, by revealing that the Space Shuttle Columbia was basically destroyed by people using Powerpoint slides attached to email, and poorly organized presentations in place of technical reports. At NASA of all places! One of the most sophisticated and technologically advanced (by necessity) organizations on the planet. And this is truly sad: people died because their support on the ground were too lazy and unsophisticated to do their work properly. Very sad.
On top of these inexcusable cases, Powerpoint can of course be used very poorly, and it is quite often. And Tufte is also right that Powerpoint’s “features” (like the “Autocontent Wizard”) encourage people to use badly. However, I think even “encourage” might be too strong a word and I would prefer to say that Powerpoint “enables” people to make bad presentations. This is what I was talking about in the first paragraph: you can use Powerpoint to make a good presentation but it’s really bloody difficult to even get it to turn out a moderately decent-looking one (and anybody who disagrees should feel free to comment).
On the helpful side, Tufte offers a good alternative to using Powerpoint: use the chalkboard and handouts along with a few slides, but don’t say “Today I’m not going to use Powerpoint.” Be subtle instead, he suggests. I’ve seen people do this and it’s really cool.
2 Powerpoint is not the problem
However at this point I have to disagree with most of what Edward Tufte had to say about Powerpoint. Tufte’s position can be summed up in his thesis that Powerpoint makes people stupid. I strongly disagree. Powerpoint enables stupidity, and even encourages it. However, saying that it makes people stupid is pushing it to the level of the propaganda that Tufte uses to satirize Powerpoint. If you see a stupid presentation, it’s because the presenter was stupid to begin with.
Furthermore, the stupidity does not result from the “cognitive style of Powerpoint.” Powerpoint is terrible because of its implementation, not because of the mental model it uses. The mental model it uses is, in my book, just fine for presenting scientific and technical data. However, Powerpoint’s pale version of it totally lacks the scientific sophistication to make someone look like a credible graphic designer or typesetter (LaTeX can do that, on the other hand). Powerpoint is a problem, but mainly because it doesn’t do what it purports to do.
That’s what I found so annoying: in Powerpoint demoting and promoting bullets and creating a reasonable hierarchy is really difficult compared say LaTeX with AUCTeX. When you’re forced into a WYSIWYG concept, you can really screw up. Powerpoint just doesn’t come close to doing it right, or easily.
Where I really disagree with Tufte, or rather think his case is weak, is that I’m a scientist and almost on a weekly basis I see presentations from people who are smart and have managed to make good presentations with Powerpoint. Yes, that’s right: good Powerpoint presentations. I wish they’d use something else, of course, but that’s not the point. The point is that Powerpoint didn’t make them stupid; it doesn’t even make them look stupid. Somebody with good data and enough time can make a good presentation in Powerpoint, regardless of how much they are limiting themselves.
A really big weak point is the alternatives that Tufte offers. With the exception of small to medium-sized meetings, where using a chalkboard is feasible, Tufte’s alternatives do not do the trick. This is not because people have been brainwashed by Powerpoint, either. His ideas are just not practical. He suggests using paper handouts. Okay, that’s fine for a lab meeting (with fewer than ten people), or maybe a medium-sized departmental meeting (at my school we have one called “Lunch Bunch” attended by 30-40 people a week). However, are you going to hand out papers at a scientific meeting, attended by hundreds of your colleagues who are loaded down with books they just bought, conference materials, abstract books and children? You see how people fill up those tote bags, man!
Paper handouts are also not practical for department-wide talks. If you put a stack of handouts by the door, the stool it’s sitting on (“Hey, can I put my coffee here?”) would get knocked over by people rushing for a seat, and nobody would bother to pick it up . People want the material to wash over them; they want to synthesize it in their heads; they don’t want to pore over tables of data.
That’s an objection to opinions he offered elsewhere in the article, too. I would venture that raw data or low-level summaries are not as effective as graphs and abstract statements: yes, people will notice when your sample size is low, but your sample size has to do with the quality of your data, not Powerpoint. Powerpoint is irrelevant.
The thing I found just totally backward was the part where Tufte invokes Richard Feynmann’s name to say “Here’s someone I know you already respect,” in a way. He points out that one of Feynmann’s books only has three hierarchical levels: chapters, sections and subsections. Well, that’s what most books have. However, he fails to mention that books like that are a step below a highly structured Bourbaki-style book written by a mathematician. Those kinds of books are easiest to read for me, because they have the most structure. The structure is clear. The hardest articles to read are the ones that have no structure below the subsection. The writer just goes on and on… Three levels of hierarchy is still a good standard, but to get the most out of a scientific text, most people need a little more. I prefer a different style.
But here’s my most compelling objection: have you seen his website? I won’t be subscribing to this guy’s ideas of design any time soon. My suggestion is to make your presentations clear and concise, using the fewest words and the most pictures possible. And to prepare those presentations, use Beamer.