On Jul 8, 2007, at 9:28 PM, Andrew Lentvorski wrote:
I would change that a little.
<feign shock/>
It's like the difference between a physicist (computer science) and an electrical engineer (programmer).
That may actually be a better metaphor, and would allow me to lump folks who work on computability theory in with those who work on string theory. ;-) The one difference is that right about now there isn't an explosively growing need for physicists like there is for computer scientists. It's like what happened with theoretical physics around the tail of end WWII.
To be a good electrical engineer (programmer) you need far less of the depth from physics (computer science), but you need a much broader base.
I'm not sure I'd agree with the broader base part. I know perfectly good computer programmers that for all intents and purposes don't know thing one about computer science. Of course, if you are good at both you the market will see huge value in what you do.
Computer science really needs to be split into computer science (mostly theoretical math) and computer programming (mostly application of algorithms).
Application of algorithms is definitely part of what this book describes as being the overt mathiness of computer science's heritage. There are huge swaths of the computer programming space that requires essentially no knowledge of algorithms. A lot of it is just writing clear code and encapsulating logic, combined with having a good sense of how to present textual information to users (think of the space of problems that 4GL's were invented to target).
I would argue that really the theoretical math vs. application of algorithms is just the usual theory vs. applied split that you usually have in the other sciences. To a certain degree this has traditionally been addressed by the differences between scientists and engineers, but of course it is somewhat more complicated than that. Either way both branches you are talking about would be about math, and math is just one incredibly useful aspect of "programming" (arguably math is an incredibly useful aspect of just about any technical field).
--Chris -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
