reflections on academia
May. 13th, 2007 03:18 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
guslA couple of weeks ago, I met with Carlos to get his advice about PhD admissions. Recommendations are most important, and publications come second. He gave me his ranking for AI schools. Surprisingly, Toronto came 3rd and MIT only 7th, behind schools like UMass, Penn and UW. He said I should be focusing on a single topic, with the goal of publishing a paper as first author, in which I propose a novel idea.
In the last months, I've been regaining much of my faith in academia, particularly the graduate admissions process and the peer-review process. If you have a good paper, there will always be a place to publish it: the system commits type II errors instead (failure to reject), i.e. errors of commission. It's high-recall / low-precision. As a whole, I think academic institutions behave pretty rationally, given the world they live in.
The explosion of information that the world has been experiencing has created not only a huge Knowledge Management problem: it also left little room for those who want to advance science in more creative ways, those who want to shift paradigms, i.e. the so-called framers. The world doesn't need more wheel reinventions. Our age has no room for rennaissance men (what about Herb Simon?), a priori philosophers. Now that all the low-hanging fruit is gone, you need to spend several years in school, climbing a narrow branch before you can reach any fruit.
This is for the same reason that we have the stereotype of the starving artist, the starving academic: everyone wants to be a framer, to work on his own ideas, and to talk about them to the world. The world already has plenty of wannabe-framers. What it needs are more fillers, so it is not surprising that this is those who it rewards.
Just as Paul Graham laments schools and suburbia as a necessary evil given the super-specialization required by the modern information society, I have come to accept that graduate schools want people who are narrowly focused on advancing their advisor's research program.
Making progress in science today requires focus and dedication: the proverbial giants, over whose shoulder we had to climb have now become a totem of giants.
In the last months, I've been regaining much of my faith in academia, particularly the graduate admissions process and the peer-review process. If you have a good paper, there will always be a place to publish it: the system commits type II errors instead (failure to reject), i.e. errors of commission. It's high-recall / low-precision. As a whole, I think academic institutions behave pretty rationally, given the world they live in.
The explosion of information that the world has been experiencing has created not only a huge Knowledge Management problem: it also left little room for those who want to advance science in more creative ways, those who want to shift paradigms, i.e. the so-called framers. The world doesn't need more wheel reinventions. Our age has no room for rennaissance men (what about Herb Simon?), a priori philosophers. Now that all the low-hanging fruit is gone, you need to spend several years in school, climbing a narrow branch before you can reach any fruit.
This is for the same reason that we have the stereotype of the starving artist, the starving academic: everyone wants to be a framer, to work on his own ideas, and to talk about them to the world. The world already has plenty of wannabe-framers. What it needs are more fillers, so it is not surprising that this is those who it rewards.
Just as Paul Graham laments schools and suburbia as a necessary evil given the super-specialization required by the modern information society, I have come to accept that graduate schools want people who are narrowly focused on advancing their advisor's research program.
Making progress in science today requires focus and dedication: the proverbial giants, over whose shoulder we had to climb have now become a totem of giants.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
(no subject)
Date: 2007-05-13 08:36 pm (UTC)The filler/framer distinction is definitely reminiscent of Kuhn's distinction between the knowledge gained during phases of "normal science" where experiments proceed more or less mechanically and methodically, and revolutionary changes. I think this tension is not particularly present (or absent) in science as a whole today: there are smaller and larger revolutions going on all over the place, and lots of "normal science" too. Perhaps we can't have "rennaissance" people who know something about absolutely everything because everything is just too big now, but the same rewards accrue to those who successfully import ideas across significant distances between disciplines. Just as a merchant must figure out what one region has and what another wants in order to turn a profit, one cannot expect to derive benefits by simply being "interdisciplinary" but only by figuring out which ideas are fruitful to import.
(no subject)
Date: 2007-05-13 08:49 pm (UTC)On your international trade metaphor, I think that AI-land has been exporting a lot to the social sciences like psych and economics, because our methods are very much in demand. Everybody wants to be a "Bayesian" nowadays.
(no subject)
Date: 2007-05-14 05:23 pm (UTC)(no subject)
Date: 2007-05-15 03:08 pm (UTC)(no subject)
Date: 2007-05-14 09:44 pm (UTC)This idea has bothered me for a long time. I would love to be a rock star of a scientist... It'd be nice not to have to invest inordinate amounts of time to get to the place where I'd have a chance of that happening. Alas the days of doing good research in your garage are probably over.
(no subject)
Date: 2007-05-16 09:04 pm (UTC)that's a nice way to put it.