[Python-Dev] Python 2.x and 3.x use survey, 2014 edition (original) (raw)

Nick Coghlan ncoghlan at gmail.com
Tue Dec 16 07:27:26 CET 2014

On 16 December 2014 at 13:08, Mark Roberts <wizzat at gmail.com> wrote:

The whole situation is made worse because I KNOW that Python 3 is a better language than Python 2, but that it doesn't MATTER because Python 2 is what people are - and will be - using for the foreseeable future. It's impractical to drop library support for Python 2 when all of your users use Python 2, and bringing the topic up yields a response that amounts to: "WELL, Python 3 is the future! It has been out for SEVEN YEARS! You know Python 2 won't be updated ever again! Almost every library has been updated to Python 3 and you're just behind the times! And, you'll have to switch EVENTUALLY anyway! If you'd just stop writing Python 2 libraries and focus on pure Python 3 then you wouldn't have to write legacy code! PEOPLE LIKE YOU are why the split is going to be there until at least 2020!". And then my head explodes from the hostility of the "core Python community". Perhaps no individual response is quite so blunt, but the community (taken as a whole) feels outright toxic on this topic to me.

The core Python development community are the ones ensuring that folks feel comfortable continuing to run Python 2 (by promising upstream support out to 2020 and adjusting our maintenance release policies to account for the practical realities of long term support), as well as working with redistributors and tool developers to reduce the practical barriers to migration from Python 2 to Python 3 (such as bundling pip with Python 2.7.9, or Brett's recent work on updating the porting guide).

It's the folks just outside the language core development community that legitimately feel the most hard done by, as they didn't choose this path - we did. Folks working on libraries and frameworks likely won't see any direct benefit from the migration for years - given the timelines of previous version transitions within the Python 2 series, we likely won't see projects widely dropping Python 2 support until after there are versions of RHEL & CentOS available where the default system Python is Python 3. In the meantime, they're stuck with working in a hybrid language that only benefits from the subset of improvements in each new Python 3 release that increase the size of the source compatible Python 2/3 subset.

Living with carrier grade operating system update cycles when you're used to upgrading your baseline target Python version every couple of years is genuinely frustrating as a developer.

Unfortunately, the anger that library and framework authors should really be directing at us, and at the commercial Linux distros offering long term support for older versions of Python, occasionally spills over into frustration at the end users that benefit from those long term support offerings.

Explanations of the overarching industry patterns influencing the migration (like http://developerblog.redhat.com/2014/09/09/transition-to-multilingual-programming-python/) are cold comfort when you're one of the ones actually doing the work of supporting two parallel variants of the language.

Regards, Nick.

-- Nick Coghlan | ncoghlan at gmail.com | Brisbane, Australia

More information about the Python-Dev mailing list