I like this conversation .. its bugged me for years that it's difficult to discuss computer science and mathematics with my friends. Indeed, I think many of us find it a bit lonely.
I certainly feel uncomfortable telling them to go get a good education in mathematics then we can chat! And when I try to explain the way GPUs and Shader languages work, I'd like to say more than "its what your graphics card does" when the real answer goes to core computer science architectures like Systolic Algorithms. And even as well as I understood "computing", taking a graduate course from Cris Moore last year let me know just what exciting and demanding ideas there are ahead. (My interest is the intersection of math and comp-sci. Much of what Knuth does.) One approach we had hoped would clarify science, technology and mathematics to non-practitioners was to create multi-disciplinary projects, first at Redfish, then at SFX. Indeed, working together with Nick on Moth (My way Or The Highway) via a netlogo model let me peer into his world a bit, and vice-versa. Many of the early SFI projects were just that: a blend of several sciences working on a shared interest. I think the down side is not Us vs Them, or "Soft science vs Hard science" or whatever. It's far simpler. It's just Damn Hard to do nearly anything of import. I think here I agree with Doug .. to really understand anything from a vortex to a GPU requires SERIOUS effort, months just to get started .. and because we only have so much time, we choose, and generally stay within our own domain because we get more done that way. Thus silos. The bad news here is that, unless I'm much mistaken Nick, asking questions won't result in much more than pointers to various techniques that have proven successful in dealing with fluid dynamics. I'd suggest we get Peter to give us a chat. I'd like some help too, now that I've worked with openGL and GPUs .. maybe http://en.wikipedia.org/wiki/Lattice_Boltzmann_methods ? -- Owen
============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
