Timeline for "Industry"/Government jobs for mathematicians
Current License: CC BY-SA 2.5
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 17, 2010 at 20:42 | comment | added | Michael Greenblatt | I did see a few funny examples of this phenomenon. My favorite was someone who got funding from the Missile Defense Agency to apply his graduate work in category theory. He made the connection from category theory in math, to category theory in computer science, and thence to computerized guidance of missiles. But mostly what I'm talking about is just very unrealistic/incompetent mathematical and statistical modeling as you described above. | |
| Jun 17, 2010 at 20:19 | comment | added | Steve Huntsman | Also I just remembered an interesting example that belies a possible reading of this answer. Hochschild was working at Aberdeen Proving Ground when "On the cohomology groups of an associative algebra" was published in the Annals (jstor.org/pss/1969145). Though it's an old example (and Aberdeen is not a Beltway bandit but part of the Army proper), I think it's safe to say that this work was not applied, and in particular was neither motivated by nor directed towards the war effort. | |
| Jun 17, 2010 at 17:56 | comment | added | Steve Huntsman | This is true. But the yardstick for defense contractors measures different things. There aren't a lot of contracts for applying étale cohomology to ballistic missile defense or what have you. Meanwhile, the LockMarts and SAICs of the world have comparatively few real mathematicians and a LOT more engineers and computer scientists (and some physicists). And since defense contracting is more "pull" than the NSF "push", somebody's got to do the work. This is why frequently you'll see silly stuff like ad hoc Bayesian methods with crummy priors and no stability analysis (this is common) in defense. | |
| Jun 17, 2010 at 17:40 | history | answered | Michael Greenblatt | CC BY-SA 2.5 |