Tuesday, April 9, 2013

The ever growing to-blog list

Something on the optimal partitioning of time based on this from Neal Stephenson and John D. Cook.

Mental Floss lists sometimes look a bit too good to be true but this list of scientific accidents would be a nice starting place for a post on the importance of open-ended research.

It's just possible that video gamers are over-represented in online polls.

The final chapter the JCP saga? We've already covered this as a fitness landscape problem, now we can talk about the dubious record of the great man theory of business management.

A whole thread on this excellent piece by E.O. Wilson on how much math a scientist really needs and this follow-up by Paul Krugman on the role of intuition.


  1. I'm not sure I'd call E.O. Wilson's article 'excellent'. The thought that went through my mind as I read it was 'senile'.

    Biology is a perfect example of a field where *more* math is becoming necessary, not less. Without mathematical training, intuition suffers greatly, and it's sad, because there really isn't the huge difference in aptitude that your average student things: you're not 'good at math' or 'bad at math'. Anyone can be 'good at math', it just takes the proper instruction and the requisite hours.

    Bias: undergraduate and graduate degrees in math before doing a PhD in statistics, so of course I think this. :)

    1. (Secondary thought: he seems to replace 'math' with 'computation' -- most of his discussion seems to be centered around computation and numbers, not mathematics. That's like saying biology doesn't really need good dissection skills, because you can always do a branch that doesn't deal with animal carcasses ... true, but beside the point. I imagine training in animal physiology is still something that most (all?) biologists share. )

    2. Wesley,

      I need to lay this out in more detail but there's a argument that's always been common in pseudo-science but which I'm now seeing in real fields like economics: using the complexity of the math to support the legitimacy of the research. That's what I'm objecting to.

    3. I agree with that, whole-heartedly. I can't lay my finger on it, but someone did a study recently where they took a crap article, added some equations, and it was much more popular (by some metric, details are hazy in my mind). Irrelevant mathematical complexity to disguise bad science should always be called out when we see it.

      I just don't see this as going as far as Wilson says it does, where he considers arithmetic to be mathematics, and then extrapolates from there. I have some friends in mathematical biology, and they're doing really interesting, ground-breaking work by using mathematical models -- work that no-one from the biology community had any hope of completing, because they lacked the tools to even attack the problem correctly.