Re: Java ready for number crunching?
All rhetoric aside, I believe that a good programmer with a mathematical
attitude could be an asset to a mathematician needing good software. The
programmer would need a mathematical background but need not be at the same
level as the domain expert.
I have been on projects with geniuses. They have much to contribute, but
their understanding is so deep that most mortals simply gape in awe,
uncomprehending. I agree with all the examples those upthread have given -
indeed, those mathematicians, researchers, engineers and other fine folks are
true masters, and I actually am in awe of them. Many indeed are good with
software - but it isn't their specialty, in many cases. Those geniuses are
well served, as they were on those projects I witnessed, by someone who could
transform those insights into workable requirements and code design. Often
this takes the form of making tools that the expert can manipulate, assistance
to their research. Other times it's to embody their design or data-structure
notions - or to embody certain equations depicting the movement of futures
markets - or to correlate the readings from widely distributed seismometers /
radar installations / satellites / experiments.
I've been on projects where programmers were turned loose on scientific data
sets to draw out more correlations in support of both static and dynamic
analysis. They really helped those scientists, and it's no slight on the
scientists whatsoever. It's added value.
Sorry for casting it so negatively prior. The ripostes to my points were
salient and convincing.