Earlier, in a rather dramatic post, Wired declared that "The Web is Dead, Long Live the Internet". Which is a pretty amusing bandwagon for them to be on in the first place.

It would be fun to rip the article itself apart but 50 odd commenters and Hacker News did all that heavy lifting.

The visualisation they used is a problem though; one that gets me regularly grinding my teeth. To any savvy reader it screams "DANGER, Unfinished Road" louder than a Looney Toons, uh, cartoon. Wylie coyote doesn't stand a chance.

The first thing that jumped out at me was the FTP traffic - which is inexplicably not registering any more. Either Wired are suggesting that no one is uploading new websites or we are exclusively using Vi over SSH for our editing1.

The issue isn't so much that the graph is so blatantly flawed; it is that it is being used to support quite a strong statement. Carl Sagan2 wouldn't be very impressed:

Extraordinary claims require extraordinary evidence
There is a bigger problem. The article is the age old apps are the future3 argument - and it builds it up by talking us through a "normal" day on the internet (which conspicuously lacks using a browser). Here's their conclusion:
You’ve spent the day on the Internet — but not on the Web. And you are not alone.
Interesting, lets look at that visualisation again; Video traffic has eaten a massive proportion of the current share. Web is "dying", Peer 2 Peer ditto. Email doesn't feature.

Trouble is, almost all of the apps that Wired list in their normal [sic] day completely fail to fall under any of these categories. Look: IM, Skype, Facebook/Twitter/News (on the iPad), RSS, Pandora, Xbox Live. Somehow I don't think that comes under "Video", perhaps Other?

Except Other is a tiny and decreasing category on the graph.

Uh oh.

Luckily they do mention "watch[ing] a movie on Netflix’s streaming service." So that's ok then4

Wired could be right about this, apps could be future, stranger things have happened. But it doesn't give you much hope when such insight is pretty much directly contradicted by the graph visualisation used to "explain".

This is a growing and annoying problem. I am sure Wired's excuse is that someone else did the visualisation (in paint and, probably, whilst feeling very clever and enlightened by the article they just read) and that the authors point stands. For the tech-savvy it's not an issue - we can roll our eyes and moan.

For everyone else it becomes a problem; because they've just read an article they don't really understand and, so, turn to the graph for support. "Look, the Web is dying, Video is the future"


Not really.

Graphs like this make people who didn't grasp the argument (at least enough to critically assess it) into mini-experts. I expect someone in a pub somewhere next week will single me out and explain to me that the web is dying5.

I mean, just say the words out loud right now - do you feel a bit silly? I did.

Graphs are dangerous in the wrong hands (lies, damned lies, and statistics) - handing one to a journalist is akin to pulling the pin out of a grenade then swallowing it. Indeed, any visualisation is like crack to them.

Data can be misrepresented to fit your theories, this is just a pain in the neck. "Data cooking" that undermines your entire thesis is just incompetence, and all the more dangerous!


  1. This is an hilarious "in" joke - I've spent all evening pretty much doing this
  2. Marcello Truzzi said it first, I prefer his version but Sagan made it quotable
  3. They probably are, right?
  4. Except it isn't, really, because you need a browser for that...
  5. Letting on you're a computer Geek attracts such attention, I've seen people swell with pride at their new found "geek smarts" when you walk into the room