• One either likes, doesn't care about or doesn't like encapsulation. I fall in the second camp so... meh.
• One either likes, doesn't care about or doesn't like functional programming. I fall in the second camp so... meh.
• One either likes, doesn't care about or doesn't like static typing. I fall in the second camp (although I admit I do have a preference for dynamic typing) so... meh.
On that third point: Whenever anyone gets into the "but muh type checks at compile time!" I just think, always, that one should write unit tests for everything anyway, no matter the language. The only difference between static and dynamic languages in that regard becomes, then, that the type in static languages is checked twice: in the test and at compile time. So I was waiting for this person to mention tests anywhere...
> Often in dynamic languages, even with a decent test suite, extensive changes can be really difficult.
And he did... with a non-point. Extensive changes are (not merely "can be") really difficult, in any language, no matter the type system.
Frankly, most of this article could be summed up as "I, really, really want Python to be Java", and the half-assed attempt at critiquing java at the end deters that notion very little. I, for one, like Python specially because it is very much not Java and I'm glad it'll stay that way.
> But I’ll be looking elsewhere if I want to write robust software.
Why do people who write this kind of articles always include sophomoric comments like this? Python has been around for longer than Java (or Go, which seems to be the other language he wants python to be, but I like Go so that's fine) and a lot of very robust stuff has been written with it.