As I'm sure you are aware, this is a parody — if one can really parody meaningless drivel — of a figure produced by Thomas Baekdal in a blog post here. PZ Myers whote about on his blog, and Mr Baekdal even left a few comments there. His First comment is #101.
Note the scale on the x-axis (it's the same in the original.)
I suggest clicking though to Blake Stacey's source. He was inspired by a merit-worthy hatchet job written by PZ Myers.
Myers target — and the original of Stacey's parody — is a deadpan history-of-information graph in a blog posting (excuse me "an article in his online magazine about 'greatness'") by info-consultant Thomas Baekdal.
I see that Grep's posting appeared while I drafted mine (and after Jens' did, oddly enough). Since he reminds me that HTML works in these comments, I'll give you Baekdal's reply to PZ Myers. He's a good sport.
@Dan Scherlis: I think my post was held up in a spam filter for a few minutes. I would load the page and see "No Comments" in large bold type right above my comment. Good thing I'm not easily offended.
Thanks for sending a link my way; I'm glad the joke was appreciated.
I'm thinking of holding a contest to find the best label for the purple stripe. Suggestions so far have included "overheard cell phone conversations", ski-lift operators, the narrator of Fight Club, Immanuel Velikovsky, Wired magazine, and Gene "Timecube" Ray.
@Blake Stacey: You missed out those important information sources "Reading between the lines", "Now they're all saying", "According to a survey" and "You know my friend, well her friend's dad knows someone who was actually there and he said".
Personally, I think it's entirely reasonable to make impressionistic claims, and to represent them graphically. Here's a place where failing to put the units on the vertical axis is the right thing to do, since it's not a quantitative plot.
I think, however, that his plot is badly wrong. Each information source is shown to trail off towards zero in influence as time passes, whereas face-to-face communication is still a large percentage of where people get their information from. I didn't learn about 9/11 from the TV or the web, I learned about it from somebody on the bus (part of "the market", I think).
@Mark F: there's also an amazing amount of stuff missing from the graph: his "careful analysis" failed to notice the existence of parents, friends, colleagues, schools, universities, books (as many commenters have noted), or sense organs! The latter are a particularly strange omission because people tend to regard information derived from direct personal experience as exceptionally reliable (reasonably enough!—though sometimes mistakenly).
The comment thread at Pharyngula contains some wonderfully entertaining rants. I particularly enjoyed the one sarcastically describing the total absence of books, letters or ships among our Cro-Magnon ancestors in early nineteenth century London.
Basically, my thought about the original graph was "Wow, that's not right at all", not "How foolish to try to make a graph like that". In fact, if he hadn't expressed his opinions about the history of information so vividly, nobody would have paid attention either to praise or reject them.
Something like 20% of all Americans get all their news from Jon Stewart's Daily Show. I'm not an American, and that is certainly my only source for "who will be the nominee for the supreme court" level US news (which tends not to be covered elsewhere.) So let's give a recentish stripe to the Daily Show.