> I would go further and argue for direct integration of numpy into the language to achieve faster performance. What do you mean by "direct integration" exactl

> I would go further and argue for direct integration of numpy into the language to achieve faster performance. What do you mean by "direct integration" exactly and why do you think it would improve performance? I initially thought you meant for it to be part of the standard library, but I don't see why that would bring any performance benefits.