I didn't get the "Python is doomed" attitude from it. To me, it's more of a rant that maintaining code that works on both 2.x and 3.x is far from optimal, and that we can and should fix this.
I think he's right in that there needs to be less of a gap, possibly with a Python 2.8. Migrating to 3 isn't the problem, it's maintaining the versions until 3 becomes dominant and 2.x support can be dropped. The Python core team is essentially deferring the difficulty of compatibility to library maintainers. And since Python 3 is essentially a new language, why not just use a py2js if/when it emerges? It'd be just as difficult. And with the additional benefit of entering a more mainstream community. So yes, let's lessen the gap.
I also agree with you in that Python 3 just needs time. And once 2.x is no longer supported, let's also remember this lesson: don't make such big leaps in language evolution.
Comments
I didn't get the "Python is doomed" attitude from it. To me, it's more of a rant that maintaining code that works on both 2.x and 3.x is far from optimal, and that we can and should fix this.
I think he's right in that there needs to be less of a gap, possibly with a Python 2.8. Migrating to 3 isn't the problem, it's maintaining the versions until 3 becomes dominant and 2.x support can be dropped. The Python core team is essentially deferring the difficulty of compatibility to library maintainers. And since Python 3 is essentially a new language, why not just use a py2js if/when it emerges? It'd be just as difficult. And with the additional benefit of entering a more mainstream community. So yes, let's lessen the gap.
I also agree with you in that Python 3 just needs time. And once 2.x is no longer supported, let's also remember this lesson: don't make such big leaps in language evolution.
The thing is, it's not even a BIG leap, just an invasive one.
This isn't all that similiar to perl5->perl6, which is a big leap.. The biggest problem is that they changed the default string type.