Wednesday, September 16, 2009

Random Musings for the Day

I like Daniel Lemire's post on "the truth" about research grants so much, I'm linking to it.

Richard Lipton is taking bets on whether P = NP; well, not really, but he thinks there are "implicit odd" in how we do research suggesting the "belief" that P = NP is in the 1,000,000 to 1 range, and doesn't see the evidence for it. (I disagree, somewhat, in the comments.)

Harvard's budget cuts have substantially closed our deficit (the "crisis" is a full further year out).

Harvard's Dean of Admissions and Financial aid answers the all-important questions. (Do I have to read now, if my oldest is a decade away from college?)

The unfolding story of the murder of Annie Le (see here if you're not caught up on your news) is disturbing on many levels, but in particular because I don't want science students -- and in particular female science students -- to feel unsafe in their buildings at whatever hours. I haven't heard much in the news reports about how situations like this (where I mean potentially unsafe working conditions such as the need to work in labs late, and -- although we surely don't know the story yet -- the appearance that the murder might have been done by a maladjusted male admirer) might impact people, and especially women, pursuing scientific careers. But the story made me think of it.


Daniel Lemire said...

Ah! Thanks for the link. It made my day...

Anonymous said...

In general, I agree with most of Lemire's article. But I take some issue with one of the suggestion that he posted (from Peter Lawerence) that: "You should be able to apply for grants based on your record alone."

Besides the obvious slippery road it puts us on, this sort of system would also make it harder for novel and interesting ideas to get funding.

Anonymous said...

"You should be able to apply for grants based on your record alone."

One can always have small starter grants for people without a record yet and/or an incredible promising idea.

It is a mistake to design a system based on the exceptional case. You deal with those separately.