Friedrich Ackermann <friedric at techfak.uni-bielefeld.de> said:
> Nobody r e a l l y fullfilling the requirements listed
> in a typical bioinformatics-job-posting (SQL, C++, Java, HTML,
> Computer Graphics, 4GL-Tools, TCP/IP, ... at least!) would do
> so for the salaries paid.
Except for SQL and 4GLs, that could describe me. (I'm more interested
in structural work, so sequence databases haven't been useful, yet.)
I don't suppose you could point me out the location of that position;
I'm looking for a job now ... preferably in the SF Bay Area :)
I imagine there are others with similar background to me. My undergrad
training was in math/physics/CS and in grad school I worked in computational
biophysics. However, I'm finding the mirror image of the question
originally posed for this thread. For my degree thesis I wanted to work
on "an infrastructure for structural biology" but there wasn't enough
physics for physics nor enough CS for CS. Additionally, it is hard to
get an advisor for this work since most of the CS people don't have much
background in modeling and the physics people don't have enough experience
in software engineering.
Continuing the thread of Larry Hunter's <hunter at nlm.nih.gov>
> I have seen the code written by some very good
> computational biologists without that kind of training. Some of those
> programs, although clearly valuable in certain contexts, would be a
> nightmare to maintain, extend, or validate.
Agreed! This is one of the chief complaints I have against a lot
of modeling programs. (Which is why I wanted to work on the
aforementioned infrastructure.) I've seen a lot of code that was written
by some very smart people, but in some cases they were too smart. They
figured out "interesting" (read "strange") ways to implement some pretty
standard data structures in a way that is very hard to understand.
This topic strikes a sore nerve in me and I could go on for a while,
but griping on that won't help things much, so I'll stop now ;)
dalke at ks.uiuc.edu