tufte

today i attended edward tufte’s seminar on “presenting data and information”.

brilliant.

it was pretty fascinating stuff, albeit i suppose you could have picked it all up by reading his four books on the subject. in fact, the course provided all four books as part of the reading materials, as he used them for reference for most of the course.

if i have to distill it down to one line (and in itself by doing this, tufte would probably be horrified— have i learned NOTHING?) it would have to be this quote:

“there is no such thing as information overload— only bad design.”

the point being that we are quite capable of digesting loads and loads of information visually. we do it every day by looking around the world, we do it by reading detailed maps. yet there seems to be this need to dumb down charts and graphs using powerpoint, which instead of making things clearer actually obfuscates things by not providing enough relevant detail.

it’s funny though; he makes a big case about how design shouldn’t detract from and can’t compensate for good content: if the data you’re presenting is crappy, no design can hide the fact. to fix this problem, get better data. which hmc points out is a nice academic attitude, often unavailable for real life. additionally, he espouses full disclosure: annotate everything, and reveal all of your data. this is supposed to give your presentation credibility, and avoid the temptation or impression that you’re cherry picking the data for the most favorable results. in fact, he describes cherry picking as “the most widespread and serious threat to learning the truth from an evidence-based report.”

and all i can think is, i’m in SALES. i’m SUPPOSED to cherry pick. i don’t necessarily want the truth. THE TRUTH IS NOT NECESSARILY MY FRIEND. i only want the version of the truth that is favorable to me at that point in time. i certainly appreciate the desire to convey all the data clearly, but sometimes that’s not necessarily my job.

sigh.

anyway, it was great to hear tufte in person, and to get him to sign a copy of his classic a visual display of quantitative information. better yet was to get work to pay for it all! w00t!

Posted at January 28, 2008 9:22 PM| TrackBack