Tufte critiques the iPhone
Tufte does a critique of the iPhone interface and there are three criticisms followed by recommendations that seem flat wrong to me. Especially since he seems to contract himself when he makes the recommendations.
I responded to his critique on his site, but just in case he doesn’t “approve” my comment, this is what I wrote:
I really have to disagree with the perspective on the weather and stock screens. You’re forgetting about, or ignoring, context and use.
The stock screens on the iPhone are clearly for tracking data, not for looking at historical data on a stock. Therein lies the difference in design. The screen is designed for just-in-time content, what is the stock doing now, what has it done today, which is appropriate. What you’re suggesting is a historical data view, which wouldn’t be appropriate for daily use. When I use the stock screen on the iPhone, I want to see what’s going on w/the market today, now, not what it’s done for the past 12 months.
Likewise on the weather screen, I want to know what it’s like right now outside. Should I grab a coat? Walk? Get in my car? What about tomorrow, what’s the weather going to be like? The addition of an animated map showing clouds moving doesn’t add value, it adds clutter. And the real value, the temperature and weather conditions, are now reduced to secondary data. It’s quite the opposite of what you say it is.
Finally, on the photo gallery—thin 1 grey pixel line instead of a 10 pixel white space? The thin grey line you propose would create almost no border between images and make them more difficult to scan through and find the right image. I’m really surprised by this recommendation, especially when shortly after you show a graphic from one of your books with t-shirts that has a great deal of white space between them. Seems contradictory.
Todd,
I attended Tufte’s course in San Jose today. He showed the iPhone video and expanded on it a little. His response to the stock page criticism you make (and apparently he’s gotten a lot of email on this) is that you *must* have that context — it’s inappropriate to look at only one day’s data if you want to get the most information from a graphic.
I think that’s debatable — some people might be able to keep the context in their heads so they don’t need it on the screen, and sometimes the task is different, like you say.
Tufte likes dense and efficient displays with everything at print/typographic resolution. He argues that adding more information doesn’t create clutter if the design is done right. But I think that assumes that you’re *reading* the document/graphic like you would text, rather than scanning or searching it (or doing whatever other visual tasks).
I hope he responds to your question. I’m sure he’s considered this and has a good argument. He’s brilliant, but I think that when he goes beyond his core expertise of statistical displays (and into, say, interface design), his ideas may need a little more thought. Not that he isn’t already way ahead of most of us.
Kevin