had an exceptional meal with my father at masala art.
found a short npr concert by dale ann bradley.
saw baltimore in a new light, as usual.
stood in the center of the fells point festival.
met the newest member of the family: james arthur dunn.
converted to the npptor text editor plug-in for the r language, a better way to use my graphical user interface of choice, notepad++. these two commands ought to be run immediately after installation:
setwd("C:\\Program Files (x86)\\Notepad++\\plugins\\APIs")
attended edward tufte's daylong lecture on the visual display of quantitative information, but skipped the autograph line.noted..
talent imitates, genius steals.
a picture should be taken as seriously as a thousand words of text.
you never need boxes. put the legend directly onto the lines and unique points. from now on, all of your linking lines should be annotated.
the metaphor for almost any efficient diagram should be a good map. maps have been around for 6,000 years, they've gotten it right.
don't dumb it down. when you're presenting to a group, those people are probably more like you than any other group of people in the world.
there is no such thing as information overload. there is only poor design.
minimize the format-figuring-out time, the format-critiquing time, the format-admiration time. maximize the content-reasoning time.
the reason you're in front of the room is a content advantage.
find a supergraphic for your presentation.
your model should be the sports section. millions of people read those tables. that's proof in the wild.
much better to have information adjacent in space rather than flip flip flip stacked in time.
use the format of the journal nature or, if you're presenting on the internet, the public library of science.
anytime you can get a real object into the room, do so. architects know this.
if you don't know how to display your data, use the small multiple.
92% of every screen has to be content.
no matter how beautiful your interface is, it would be better if there were less of it.