search

the-teachingRSE-project / competencies

The teachingRSE project: "Teaching and Learning Research Software Engineering"
Creative Commons Attribution 4.0 International
27 stars 19 forks source link

legacy RSE are doing COBOL #324

Closed mhagdorn closed 1 month ago

mhagdorn commented 1 month ago

is there really any research software that uses COBOL?

CaptainSifff commented 1 month ago

Good point.... any alternatives? Algol? Pascal? APL? Lisp?

mhagdorn commented 1 month ago

I think Pascal might be a better one.

MakisH commented 1 month ago

Remember that language standards also have versions and this could be an easy way out.

Legacy Fortran (meaning anything else than modern Fortran, e.g., Fortran 77 and maybe Fortran 90, would be a clear candidate with many examples.

In the same category, I would also count (older) C and C++ pre-C++11 as legacy (but I don't know to what extent that could backfire). I definitely don't want to work with these nowadays if I have other options.

I also think it looks strange writing about COBOL in a research context.

CaptainSifff commented 1 month ago

hmmm.... so let's not state that we're so old, that we know these languages and instead do sth. like: "legacy code written in Python-2 or pre-Fortran-77 code"?

mhagdorn commented 1 month ago

yes, that's much better. I'd say pre-fortran77 is truly ancient, what about pre-Fortran95. F95 is when the language gained a lot useful features that made it quite modern.

MakisH commented 1 month ago

Maybe something like:

code written in older language standards, such as Python 2, Fortran 77, C++98, or other language standards considered deprecated by their communities.
knarrff commented 1 month ago

I am not against it, but putting Python 2 into the same basket as Fortran 77 seems a little strange to me. There are 30 years between these two, while one of them is only 16 years old. Python 3 was released 2008. To be comparable, we would have to use Fortran 2008 - but then of course that's still "new" to the Fortran community.