5 ms·
Some of the Numpy and Scipy routines go back earlier than the 90's. AFAIK, they trace their roots to numeric routines on Fortran punch cards.
by MengerSponge 11y ago
Some of the Numpy and Scipy routines go back earlier than the 90's. AFAIK, they trace their roots to numeric routines on Fortran punch cards.
- masklinn 11y ago> AFAIK, they trace their roots to numeric routines on Fortran punch cards. Oh yes they can bind to much older libraries (e.g. BLAS), I was talking about the packages themselves. For instance Numpy is the unification of Numeric and Numarray, the former having gotten started circa 1995[0] on matrix-sig[1] and the latter originally indented to replace the former. So numpy itself has its roots in 1995. [0] https://mail.python.org/pipermail/matrix-sig/1995-August/000002.html https://mail.python.org/pipermail/matrix-sig/1995-August/000... [1] https://mail.python.org/pipermail/matrix-sig/ https://mail.python.org/pipermail/matrix-sig/ There are gems in these early messages too, for instance one of the early threads suggested allowing parens-less tuples and slice literals (a:b) in tuples within indexing: https://mail.python.org/pipermail/matrix-sig/1995-September/000044.html https://mail.python.org/pipermail/matrix-sig/1995-September/... Slice literals, let alone slice literals in tuples, remain indexing-only features to this day. So the language has a long history of adapting itself to numerical computing concerns.