breakfast first

Shame on the Waterfront

I'm ashamed to admit that, until yesterday, I was unaware that Yo La Tengo was from Hoboken. A lot of interesting things come from that tiny city. (I've been obsessively listening to "Autumn Sweater".)

Just not getting it

I typically like a lot of what Seth Godin writes, but his post today is the most ignorant thing I've ever heard him say. In the post, he says "Pythagorus [sic] and his followers...didn't realize that they should have been working on calculus instead of messing around creating and polishing stuff that no one needed or cared about [specifically, perfect numbers]."

Seth, that's the whole point of pure mathematics -- to survey the landscape, find connections, and add to our knowledge of mathematics as a whole. Applications may come later, or may not come at all. That's fine. You can't know if it will be useful unless you've been there and explored. Your statement about studying perfect numbers is akin to saying, "Why waste time surveying a mountain if we're never going to build a mall there?"

There are thousands of stories about math research that at first seemed "useless", only to have amazingly important consequences and applications later on. Number theory, which includes the study of perfect numbers, seemed pretty stupid -- until cryptography came up. Without people having studied prime factorization, you wouldn't feel safe making purchases over the internet.

More near and dear to my heart is knot theory, which I studied in graduate school. It's essentially the mathematical study of how to determine if two knots are the same. Could anything sound more dumb? People worked on this, without any application, for a hundred years. And then suddenly, biologists realized it was very useful for understanding DNA replication (among other things). Dang, didn't see that one coming.

So come on, Seth. Pure research gets a bad enough rap from the American Idol-watching public without having a smart guy like you only thinking short-term.

A Weird New Metric

Someone recently told me that, when looking interviewing with a company, he looks at their LinkedIn ratio: the number of current employees registered with LinkedIn vs. all employees registered for that company, past and present. Even though that contains severe sample bias (since people switching companies are more likely to register with LinkedIn), he still thought it was useful. I'm unconvinced, but it's an interesting use of the service.