Retiring old Fortran code for linalg.interpolative and possibly sparse.linalg.iterative

Dear all,

We have been discussing the fate of the mentioned codebase, in the RFC [1]

We are coming to a point where it looks like a semi-consensus that we need
to do something about them. Less with the "sparse.linalg.iterative" which
only has open issues burden but mainly "linalg.interpolative" which is
causing all kinds of compilation issues.

Just to mention, I think we don't have any appetite left for discussing the
state of Fortran (multiple times already and also see [2] ) moreover Ondřej
Čertík already broke some really good news about LFortran project. So let's
try to get to a decision about scipy code first then we can digress
further.

The rather interesting option; can we deprecate "linalg.interpolative"
instead of modernize? Input for that would be great.

Thanks,
ilhan

[1] : https://github.com/scipy/scipy/issues/18367
[2] :
https://discuss.scientific-python.org/t/releasing-or-not-32-bit-windows-wheels/282

_______________________________________________
SciPy-Dev mailing list -- [email protected]
To unsubscribe send an email to [email protected]
https://mail.python.org/mailman3/lists/scipy-dev.python.org/
Member address: [email protected]