Not a dumb question at all. I left my Tufte books at the office, so I can't check, but I thought he cited sources throughout.

Either way, between Cleveland, Robins, Few, and Ware there're tons of studies and references:

http://cm.bell-labs.com/cm/ms/departments/sia/wsc/index.html

http://www.researchgate.net/researcher/2028337238_Naomi_B_Ro...

http://www.perceptualedge.com/library.php

http://www.amazon.ca/Visual-Thinking-Design-Colin-Ware/dp/01...

Better is subjective though and related to what you're trying to do. A lot of past research was about efficiency of information transfer. There's more recent research looking at memorability. In one report, different graphics could have both goals.

There's a "Useful Junk" paper delving into that topic. There's a link to it in Few's reply: http://www.perceptualedge.com/articles/visual_business_intel...

There's plenty more. Stanford Vis group is a fun place to check from time to time: http://vis.stanford.edu/ Heer specifically has a range of papers and some online slides for a vis course he taught http://homes.cs.washington.edu/~jheer/ (uses Tufte for the course text)

I'm drawing a blank on a few others I find informative...

Unfortunately, I don't have any about fonts but I'm sure there are plenty out there.

One of the studies we were shown in a usability class was that pretty interfaces are rated by users as more satisfying and actually are better at letting them accomplish goals as well. If you went to a really ugly web site with lots of clutter, like the myspace era, and got a progress dialog, for example, you might just ditch the web site. Whereas if you went to a beautiful web site and it gave you a progress dialog, you might be willing to give it the benefit of the doubt, for example. Sort of like how tall, good looking people earn more money. Sad, but factual.

Right, many software devs have read The Design of Everyday Things [0] and understand the value of usability. But they didn't read Don Norman's follow up book Emotional Design [1] which explains that usability is only 50% of the answer - emotion is the other 50%. And emotion often means pretty, or at the minimum a positive UX.

[0] http://www.amazon.com/Design-Everyday-Things-Revised-Expande...

[1] http://www.amazon.com/Emotional-Design-Love-Everyday-Things/...

I wouldn't be surprised to find that emotion is actually higher than 50%. Though, I like to gripe about how overblown the study of doors is in that work.

One interesting aspect of that is progress bars. Regardless of the amount of time an operation takes people will have a slightly different subjective opinion of how long it actually took depending on the behavior of the progress bar. Progress bars that slow down near the end seem to take longer than progress bars that accelerate near the end, even if the elapsed time is exactly identical.

To focus on an issue that Tufte often addresses, pretend that you are Lawrence Mulloy and it is late on the evening of January 27, 1986. Several engineers from Morton Thiokol have called you and told you that if the Space Shuttle launches tomorrow morning, it will explode, and all of the astronauts will die. But it is expensive to call off a launch, so you ask them what data they have.

Would you say "launch" if you only saw the first 2 charts on this page:

http://www.asktog.com/books/challengerExerpt.html

Would you say "launch" if you saw the 3rd chart on that page?

The decision to launch tomorrow is yours. What do you decide?

That is the most ridiculous curve fit I've ever seen.

I can't see any of the charts on that page. They're just collections of dark splotches.

i suggest you read one of tufte's books - the classic is the visual display of quantitative data. he dissects 'common' charts etc and then strips away the baggage - that's pretty much what it means. what you call 'arty' is (maybe) some icing on the cake but the core is pretty solid.

I've read most of Tufte's books. And I tend to agree with them 90+% of the time.

But anecdote is not the plural of data. Pretty is not automatically effective.

Nielsen Norman Group occasionally hit some unusual exceptions when they actually tested their assumptions about web usability even though they got most of it right. I suspect the same with Tufte.

>the classic is the visual display of quantitative data

...I've never read any of his works (I'm not the OP), is that the best Tufte book to start with?

That or "Envisioning Information." "Visual display..." is slightly more quantitative (as the name would suggest). He has some ebooks for $2 if you just want to get a feel for his writing and design: http://www.edwardtufte.com/tufte/ebooks

If you want science, read Graph Design for the Eye and Mind[0] instead.

[0] http://www.amazon.com/Graph-Design-Mind-Stephen-Kosslyn/dp/0...