Electric Josh

How to count crowds
This sort of thing is done all the time, and shouldn't be hard. Take a bunch of pictures of all parts of the crowd. Better yet, have a news helicopter scan the whole crowd from above with a video camera and then stitch together a single, big image using some clever image processing. Then digitally sign the image and post it to a web site that both the insane left and the insane right can download it and do their own analysis. Since the picture (or set of pictures) is digitally signed, nobody can forge denser or sparser crowds. Because everyone can see the same picture there's less opportunity to fudge the analysis, or to just make up numbers. It's certainly not nearly as hard as exit polling.
[permalink | tb ]

1, 2, 3,...
So the vigilant Tom Tomorrow alerts us from his usual place in the sidebar of this account of the MLK weekend antiwar march in Washington. If it is to be believed, it would seem that media outlets are reliably, systematically undercounting the size of antiwar marches. Damn liberal media.

Of course, my excellent (formerly Serbian) colleague points out that that's a pretty small protest. "Do you mean you've heard that it wasn't very big, or are you saying that a quarter million people is 'pretty small'?" I asked him. "Yes. It's a tiny fraction of the voting population. Even 250,000 people is pretty small compared to protests that actually get something done."
[permalink ]