Beautiful Evidence, by Edward Tufte

It’s a beautiful book, of course. Tufte doesn’t really do anything that’s not incredibly well-designed and constructed. This is his 4th book on, um, information design, I suppose you’d call it. Read it over the weekend, in a bit of a return to reading about my field (HCI) for a while. I’ve not read much on design in a long time — I think I got sort of fatigued for a few years — but now have read Maeda, Tufte, and a few other things in the space of just a few weeks. I’ve got more queued up, including Moggridge’s compendium. One thing I learned this time around is that Tufte self-publishes all his books. That’s why they’re perfect, I suppose, and why they’re harder to get than books that go through what I’d call more normal channels.

So it’s a good book. But feels a bit like a mishmash of greatest hits (could he put that map of Napoleon’s trip in more books?) mixed together with a rant on how Powerpoint is killing all that’s good in this world. His point in this book, more than any of his others, is that if you’re careful, you should always be able to let evidence speak for itself. You don’t need to market it. You don’t need to spin it. Just present it in ways that are meaningful and you’re there.

Part of it, I suppose, is that I read about & internalized the idea of sparklines a pretty long time ago. Here’s the chapter — very very worth reading. They’re wonderful. Every designer I know has been looking for some practical use for them, which has proven elusive. Maybe here’s one that would work. (Although I wonder if there’d be an effective way to get enough datapoints to be useful.)

His point about Powerpoint is that Powerpoint makes us dumb. That the structure of the software necessarily makes us reduce the amount of meaningful information that we’re presenting, and ultimately make arguments which, without evidence, aren’t compelling. Both absolutely correct and beside the point, of course. This essay has been critiqued elsewhere, so I’ll not do that again (although I very rarely voice quite the same opinion as Don Norman), but i will say that the whole chapter — maybe 1/4 of the book was devoted to it — felt deeply out of place and not very relevant to me.

Anyway, like a lot of expensive design books, this one is probably better for borrowing than for owning. Tufte has said there’ll be a 5th book in the series — if history is an indicator, we should expect it in 2013 — so hopefully it’ll be a great finish to an important series, and everyone won’t worry too much about this one.

4 comments

  1. He’s doing his one-day course in SJ in early December. Dolske and Faaborg are going, and if you’ve never been to one and can spare the day, you should think about it. He’s a great speaker, and I found the presentation to be a lot richer than his text. Also, you can see first-hand just how eccentric and devoted to data visualization the man is, which is fun.

  2. Oh, and, uh, I’d borrow that book from you …

  3. He’s doing his one-day course in SJ in early December. Dolske and Faaborg are going, and if you’ve never been to one and can spare the day, you should think about it. He’s a great speaker, and I found the presentation to be a lot richer than his text. Also, you can see first-hand just how eccentric and devoted to data visualization the man is, which is fun.

  4. Oh, and, uh, I’d borrow that book from you …