As someone who has been both a Perl and a Python programmer, I don't feel like Python3 is going to kill Python. Why?
- Python3 currently exists, and there is a large push to get the majority of the "big" packages to support it (many of which do currently support Python 2.x and Python3).
- It's possible for a Python package to support both Python 2.x and Python3 at the same time or to program the Python 2.x version in such a way that converters like 2to3 can do the conversion for you. I'm aware of no such tools or capabilities between Perl 5.x and Perl 6.
- Python3 isn't as ambitious as Perl 6 is/was.
Edit: As to point #2, I recall that there may have existed (maybe it still does?) some project to use Perl 6's "reprogram the grammar" capabilities to turn Perl 6 into Perl 5. I'm not sure how I really feel about that though. There was also a time when there was some buzz around Parrot VM[1] that was associated with Perl 6 (I don't know if it was some official partnership or just some announcement of Perl 6 support in Parrot).
No. Python3 delivers concrete improvements on accepted problems with python2. Just the async io and encoding improvements with python 3 justify it. It continues the design spirit that drew devs to python.
When a loved platform stops evolving, as python2 has, there is this lovely period when it continues to be awesome, while requiring you to learn nothing new. Python is here now. But come the day that my new Linux distro links to 3 over 2, I will get organised to move.
Not at all. Unlike Perl 6, Python 3 1) exists 2) has a significant amount of library support and 3) is clearly intended to supplant Python 2. None of those points apply to Perl 6.
No, it doesn't seem like the same issue at all. I used to use Python for most (production) things, between 2003 and 2008. Due to work, I more or less switched over to PHP and Java, and hadn't written any new Python in anger in years until a month ago. Despite all that, doing some simple stuff in Python 3 felt exactly like Python 2 always had. I'm sure that people who use it on the daily will notice and stumble against lots of differences, and porting large codebases is hardly ever fun, but it's fundamentally the same language.
I would think that the PHP4 -> PHP5 (relative) slow transition was more about hosting support than developer's adoption. Because of of the extreme (and often infuriating) stance on backward compatibility migrations were not that difficult : http://php.net/manual/en/migration5.incompatible.php
It was mostly new things, not different things. I don't even remember the new object model being that hard to pick from the PHP4 one. I think lots of developers got more stuck on the 5.2 -> 5.3 transition.
Isn't Python 2/3 more about libraries and compatibility than hosting ? My personal anecdote is that it deferred me way too long to learn the language seriously. I didn't want to waste time learning an old version but too much things were on it, especially Django and Flask (being mainly a webdev).
- Python3 currently exists, and there is a large push to get the majority of the "big" packages to support it (many of which do currently support Python 2.x and Python3).
- It's possible for a Python package to support both Python 2.x and Python3 at the same time or to program the Python 2.x version in such a way that converters like 2to3 can do the conversion for you. I'm aware of no such tools or capabilities between Perl 5.x and Perl 6.
- Python3 isn't as ambitious as Perl 6 is/was.
Edit: As to point #2, I recall that there may have existed (maybe it still does?) some project to use Perl 6's "reprogram the grammar" capabilities to turn Perl 6 into Perl 5. I'm not sure how I really feel about that though. There was also a time when there was some buzz around Parrot VM[1] that was associated with Perl 6 (I don't know if it was some official partnership or just some announcement of Perl 6 support in Parrot).
[1] http://www.parrot.org/