what software do biologists need?

Ben Jones jonesbb at BELOIT.EDU
Mon Mar 22 22:32:30 EST 1993

Here's another request for a software package:  Software for mapping
migration patterns on a geological map or globe, using mark and recapture
data.  It should be able to distinguish live sightings from deaths, and be
able to divide the data up by distance from the marking site, by date of
marking, and by date of observation.  It should be able to estimate and
display probable migration routes from the data.  It should be able to use
scanned images for the maps, and ideally to use maps from commercial map
software libraries.  A wishful thought would be to correlate the patterns
of migration with geological features, ocean currents, wind patterns etc.

Raul Valdes-Perez writes:

>The biological relevance is that asking biologists what software they 
>need misses a lot of interesting opportunities.  The computer scientist, 
>who after all should be the expert in what is practically computable 
>(although perhaps as a research problem), should spend part of his time
>trying to dream up things that biologists never thought could be computed,
>because that's not what they are experts at.  Of course, the computer
>scientist has to know something about biology to even get an idea.

I think I agree only with the last sentence, and even then its an
understatement.  Computer scientists may dream up all kinds of things that
biologists never thought could be computed, but it won't often be something
useful to a biologist.

Programmers with a computer science degree generally know nothing about
what biologists need.  My experience is that that in order to come up with
good software for any given field, including non-science fields, it is
essential first to know *that field* well, and secondly to also know the
capabilities of computers.  It is not necessary to have a computer science
degree.  It is not essential to even know how to program, because that is
merely a skill which is easily learned by people comfortable with
computers, and anyway you can hire a computer scientist for the grunt work.
 You don't necessarily need to know the latest, greatest algorithms, nor
the newest cutting edge technology, though they may help later.  What you
need is the imagination to know what tasks can be computerized in such as
way as to make them easier and faster, or even possible at all.  And maybe
later you find that additional tasks suggest themselves, or other
biologists suggest things, or computer scientists suggest things.

(And then you need time and/or money.  Lots of it.  And tenure. :-) )

In terms of the bridge analogy, you build a little footbridge for yourself,
and maybe discover that other people start crossing it.  And maybe somebody
gets to the other side and discovers you can build a tower there.

Ben Jones                  BioQUEST / Department of Biology
jonesbb at beloit.edu         Beloit College, Beloit, Wisconsin

More information about the Bio-soft mailing list