5 ms·

The author does acknowledge that COBOL and FORTRAN are still around in the ALGOL section: > Of the four mother languages, ALGOL is the most “dead”; Everybody s

The author does acknowledge that COBOL and FORTRAN are still around in the ALGOL section: > Of the four mother languages, ALGOL is the most “dead”; Everybody still knows about LISP,2 COBOL still powers tons of legacy systems, and most scientific packages still have some FORTRAN. Which is kinda true. COBOL and Fortran are not "dead", but they're heavily "deprecated"; I don't think there are too many systems created today in COBOL vs just maintenance and support. Maybe it's not the same for Fortran: it still powers a big chunk of Python's scientific stack (in the scipy package).

pnt12 4y ago
Interesting, I assumed all high performance packages for science and math either used c or c++! Is there a reason to use Fortran instead?
- detaro 4y ago
  Fortran has enough critical mass in that field. I.e. good libraries and applications are written in Fortran, so enough people in the field learn Fortran and use Fortran instead of spending time rewriting stuff that works and is well-tested in C++.
  - naasking 4y ago
    It's also that there is no recognition or other incentives for developing or maintaining software in these fields. There was an article a week or so ago about a growing panic in particle physics over an important C program maintained for 40 years by a single prof who will soon retire.
- wglb 4y ago
  From the standpoint of an optimizing compiler, Fortran is much easier to optomize. The existence of pointers in C make that much more difficult. As Francis Allen said (paraphrasing) "The prevalence of C has set back programming language research by decades."
wglb 4y ago
> COBOL and Fortran are not "dead", but they're heavily "deprecated"; I don't think there are too many systems created today in COBOL vs just maintenance and support. I don't get the sense that they are except in some segments of programming. Do you have any reference that would show that they are deprecated? And Fortran is much bigger than scipy.
- santiagobasulto 4y ago
  I do know that Fortran is a lot bigger than Scipy, I was just mentioning that it's "so important and relevant (still)" that it powers the whole "Data Science" Python world. I have no sources for my claim, just complete gut feeling. I'd bet my lunch that there are more new systems started in Java than COBOL. But I'm happy to be proven wrong and fast until dinner :)

pnt12 4y ago

Interesting, I assumed all high performance packages for science and math either used c or c++! Is there a reason to use Fortran instead?

detaro 4y ago
Fortran has enough critical mass in that field. I.e. good libraries and applications are written in Fortran, so enough people in the field learn Fortran and use Fortran instead of spending time rewriting stuff that works and is well-tested in C++.
- naasking 4y ago
  It's also that there is no recognition or other incentives for developing or maintaining software in these fields. There was an article a week or so ago about a growing panic in particle physics over an important C program maintained for 40 years by a single prof who will soon retire.
wglb 4y ago
From the standpoint of an optimizing compiler, Fortran is much easier to optomize. The existence of pointers in C make that much more difficult. As Francis Allen said (paraphrasing) "The prevalence of C has set back programming language research by decades."

wglb 4y ago

> COBOL and Fortran are not "dead", but they're heavily "deprecated"; I don't think there are too many systems created today in COBOL vs just maintenance and support. I don't get the sense that they are except in some segments of programming. Do you have any reference that would show that they are deprecated? And Fortran is much bigger than scipy.

santiagobasulto 4y ago
I do know that Fortran is a lot bigger than Scipy, I was just mentioning that it's "so important and relevant (still)" that it powers the whole "Data Science" Python world. I have no sources for my claim, just complete gut feeling. I'd bet my lunch that there are more new systems started in Java than COBOL. But I'm happy to be proven wrong and fast until dinner :)