[[clean-list] Clean and scientific programming]

[Olivier Lefevre]

Siegfried Gonzi <siegfried.gonzi@kfunigraz.ac.at> wrote:
> Nobody knows why [FORTRAN, C++, Perl, Python etc] can attract so 
> many people and especially developers from universities.

You are laboring under the illusion that academics engaged in 
scientific and/or computational research will use the best tool for
the job. But the reality is that they often don't do a thing besides
writing grant applications: students (including postdocs) do all the
programming. The students, OTOH, know there is a high probability
that they will have to make a living as programmers of general
business applications and therefore push for the use of languages
with well-established credentials in the general computing community.
Meanwhile their advisers don't care because either they don't know
better or they do but, heving access to a bottomless supply of free
or very cheap labor, they have no incentive to look for productivity
gains by using betetr tools: if the project does not progress fast 
enough, they can just throw another grad student at it.

This is of course similar to the situation on the hardware side,
where scientific computing is too small a market to drive architecture
development anymore and scientists have to make do with is available 
for business users.

For these reason I, for one, am very skeptical that the scientific
community can be a good launching pad for anything and I rather 
approve of the decision of the Clean team to focus on things like 
games.

Regards,

-- O.L.