Over the last few months, I've read several books and articles that concern the graphical presentation of data and information using presentation software, such as PowerPoint and OpenOffice.org Impress (both of which I have used a lot) and Keynote (which I have never used. I am approaching the problem of presentation from the angle of making scientific presentations, often with considerable quantities of complex data.
Edward Tufte on the cognitive style of PowerPoint
I've reviewed a couple of Tufte's books on graphic design elsewhere on this site (e.g. here and here). He has published a chapter on PowerPoint (excerpted from Beautiful Evidence), entitled The Cognitive Style of Powerpoint: Pitching out Corrupts Within. There, Tufte sets out in typically trenchant style his objections to PowerPoint as a vehicle for communication of information. He certainly supplies examples to support his contention: a couple of PowerPoint slides from the enquiry into the Challenger space shuttle disaster inquiry, and a presentation of Lincoln's Gettysburg address among others. He objects to the use of multiple bullet point levels, typographical inconsistency and the minimal data display (in part due to the low resolution of the screen versus the printed page). It is with this last point that I feel his argument falls. PowerPoint presentations are not documentary media, and if they are used as such this is inappropriate usage. Tufte presents a table of median number of data matrix entries for statistical graphics in a variety of publications: Science tops the list with >1,000, while books about PowerPoint presentations come second to bottom (12 data entries), beating only a 1982 issue of Pravda (5 entries). But here, Tufte is not comparing like with like. In delivering a presentation, one really only shows an overview of the detailed data - if the details of the data are important, then these are indeed best supplied by the printed medium.
The peril of the template
Where PowerPoint becomes a problem is in the use of the predefined templates, and to a slightly lesser degree, the slide layouts. Indeed one of the liberating features of merely presenting sequences of images or pdf pages via KeyJNote (not to be confused with the Mac application Keynote) or similar applications is that one is freed from the temptation to drop in the template or slide layouts. Kosslyn (Clear and to the Point: 8 Psychological Principles for Compelling PowerPoint Presentations) demonstrates the perils of the template, and discusses the psychological pitfalls into which a presentation designer can fall. Kosslyn makes an argument for limiting the amount of information within each slide. Incidentally, the eight principles all make very good sense. They are, though, considerably less radical than the presentation style espoused by Garr Reynolds.
Garr Reynolds and Presentation Zen
Garr Reynolds runs a rather neat blog on presentation skills, and has published a nice book on the same topic, Presentation Zen. Reynolds is a very experienced presenter, and espouses a particular style in which the presenter assumes a much greater importance than is usual in PowerPoint presentations (at least in the technical talks and lectures that I usually attend). Nonetheless this is a refreshing approach, albeit slightly Mac-fanboy in tone. (I am more aligned towards Linux in my PCs. Actually, I'm just jealous about the physical beauty of Mac notebooks and some of the software packages. But I'm just too mean to cough up for it!).
Reynolds' slides are visually arresting, but unfortunately his style isn't directly applicable to technical presentations. That's not to say that those of us giving research talks cannot benefit from a dose of Presentation Zen makeover - actually, I believe his stylings have had a direct impact on the way I currently design slides.
What software do I use?
I can't use Keynote, as I don't have a Mac. I do use PowerPoint, though less frequently these days as I move away from Windows towards Linux. I recently used KeyJnote for a presentation (Ageing - Why do we age and what can we do about it) at a local secondary school. While I found presention sequences of images quite liberating, I found the process of designing ordering and organising the images slightly awkward. It is the facility for rearranging the playing order and modest animations (very useful for explaining chromosome segregation!) that makes presentation software so useful. My favourite is currently the OpenOffice.org presentation component Impress. I don't touch the supplied templates, and I only use the blank slide layout. Using only the blank slide layout, with no predefined text, table or image placing, frees me to generate what I feel are more pleasing slides.
Is PowerPoint quite as bad as its critics claim?
No, I don't think so. It has the facility to encourage bad presentation habits, but we don't criticise MS Word for enabling some real abominations of font mixtures and layout. If people can be persuaded not to use a lot of the fancy stuff such as busy templates, over-complex slide layouts, ugly slide transitions and wild animation effects, it is an excellent tool. Most of the issues people raise are operator issues.
So some of Tufte's complaints are valid (but are due to the presenter's (lack of) skill at slide design rather than an intrinsic failing in the software. His issue with data density is rather mistaken I believe in that he requires presentation software to do something inappropriate.
I think that we can see the management guru styles of Reynolds and co as one extreme of the presentation spectrum, with the hard-core science presenters at the other, though the need to make presentations as accessible and engaging exists for both.
Other links
Scott Schwertly's Presentation Revolution website has many tips - his manifesto is well worth looking through.
Ian Parker wrote an amusing critique of PowerPoint, published in The New Yorker (2001). It includes a bit of PowerPoint history.
Seth Godin's pamphlet Really Bad PowerPoint (and how to avoid it) takes a few potshots at PowerPoint pitfalls, and makes some interesting points.