Re: [tlug] Fortran --> Python (was linux engineer)

[BALUTA Chris <baluta@example.com>]

   Since I know nothing of the biologial sciences, chemistry, etc.,
 I'll restrict my comments to the physical sciences (astrophysics,
 in particular, though I don't think physics is terribly different).

   The software used in data analysis, whether it be in-house or
 (for  very large observatories - think Hubble) a package/suite
 of tools provided to the community at large, might be written
 in any number of languages. For most of what I have used/seen -- and
 this fits into the 'tools provided to the community' category -- the
 'guts' of the software has been written in Fortran, C, C++ or IDL[1]. 
 Perl, and of late, Python, come into the picture as scripting languages: 
 "I need to run seven or eight tools on my dataset in order to extract 
 a spectrum" == "I am going to write a script". (Typically, the software 
 packages provide such scripts.)  One exception to this: the Herschel 
 space telescope (an EU/ESA infrared  observatory) has done their analysis 
 software in JYthon. 
 
   I suspect that where Python is gaining traction (other than in the
 scripting case mentioned above) is under the 'in-house' category - a 
 graduate student needs to do some fairly light numerical work, and codes 
 up a Python script. It's not uncommon to find links to such software 
 under the category of "Non-standard analysis tools", for example.

   As for the question "Why (still!) Fortran?": When I was in graduate
 school (20 years ago...sigh), we had a collaborator from Germany
 who constructed state-of-the-art stellar atmospheres. I asked him how 
 big his (fortran) code was. "When I was in grad. school it was a few 
 million lines...I haven't counted since then.")  So... Who is going 
 to re-write a few million lines of Fortran code into Python? (And why?!)
 It's a good bet that graduate students studying under him did not write
 their own stellar atmosphere code in perl or python or c or ... And
 so with the students studying under them. (I've also heard from some 
 theoriticians - even 'young' ones - that  Fortran is still the superior 
 language for numerical work.)



 [1] No, not "Interface Description Language", but "Interactive Data Language".