Friday links: “new” Alan Turing paper, and more

Also this week: evidence that hindsight is indeed 20-20!

From Jeremy:

Stephen Heard on how he got into blogging. His Scientist Sees Squirrel is the best new ecology/science blog in a long while. If you’re not reading it–well, why the heck not? Interesting to hear that Stephen got into blogging through writing a book. I’m hoping to go in the opposite direction…

The ESA now has a formal code of conduct for their annual meeting.

Often in science, you’ll get a surprising result. It’s a much bigger effect than you were expecting, or in the opposite direction to what standard theory would predict, or whatever. And often, it’s possible to come up with a post-hoc explanation that makes the result seem unsurprising (or at least less surprising) in retrospect. But as Andrew Gelman reminds us, on its own that is a highly unreliable procedure. Because it’s very easy to come up with a plausible-seeming post-hoc explanation for anything. Even fake data that were designed to be surprising. And I’d add that, now that the data have been revealed to be fake, a lot of people are saying that someone should’ve recognized the fraud even before it was published. Which is unfair, and which provides a second illustration of Andrew Gelman’s point. Never forget: everything is obvious once you know the answer. I’ll also add that I disagree with a lesson Andrew seems to draw (not sure if he intends to draw it, though others have). Namely that one should have blanket skepticism about anything published in the highest-impact journals (Nature, Science, PNAS). At least in the fields in which I have expertise, Nature, Science, and PNAS mostly publish good work that doesn’t ring any alarm bells. That the top general science journals publish some flawed papers, or that they (may) publish a greater fraction of flawed papers than more specialized journals (as is sometimes claimed even though it’s not demonstrated by the evidence everyone likes to cite), does not justify dismissing everything they publish out of hand. (Aside: further details about the fakery incident that prompted Andrew Gelman’s comments here, here and here. The demonstration that it is indeed fake is here [I read it, it’s devastating]. A bit of sensible commentary here. This incident is all over the intertubes, so you won’t have to look far to find further commentary and speculation of varying levels of sensibleness.)

Alan Turing has a “new” preprint up on arXiv, concerning the application of probabilistic reasoning to cryptography. I say “new” because it was written during WW II but only recently declassified. HT Andrew Gelman, who has some good comments on Turing’s reasonableness and good judgment. I agree with Gelman that good judgment is a very underrated trait in science. A lot of what Meg, Brian, and I write about is our own hopefully-good judgments about all sorts of stuff.

4 thoughts on “Friday links: “new” Alan Turing paper, and more

  1. thanks for the links! FYI, some words seems to be missing in the middle of the long paragraph (“. should have blanket skepticism…”)

  2. “I’m hoping to go in the opposite direction…” Yes, I suspect that’s an ultimate goal for a number of us blogging in science. Deciding what to leave in, what to re-write and expand, and how to structure the final product is certainly something that I’m struggling with at the moment.

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s