the existence of numpy seems like the basis of python becoming the language for data science.
was there anything about python that made numpy easier.
could there just as easily been a numpy for ruby?
@fgregg did Notebooks predate NumPy? IPython is just a few years older than the NumPy name but I want aware enough in 2001-2005 to know what was drawing people into Python at the time. Ruby certainly didn’t feel real to me until Rails in ~2006; I’m unclear why it was chosen by a PHP refugee.
@urschrei @fgregg If 2011 is the beginning, then it’s after all the data science “big three” of NumPy, IPython, and Pandas. Sometime around there you already had Software Carpentry attracting philanthropic funding so it must’ve been clear that this Python ecosystem was already winning and should be supported.
@migurski @urschrei @fgregg Not as a primary cause, but contributing, I think, is that by 2007 python has a reputation for having a library for everything: https://xkcd.com/353/
@vruba @urschrei @fgregg PG was also bloviating about it as early as 2004 https://www.paulgraham.com/pypar.html