Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Sunsetting Python 2

python.org

Post reply on HN
Print view View original

251–260 of 733 posts

Re: Sunsetting Python 2

#251

Post by Mikhail_Edoshin »

simias wrote: post #182
brobdingnagians wrote: post #63

Earlier quoted context omitted.

Tongue in cheek comment; C was released 47 years ago and is still supported... but then again, that shows a very different philosophy between system languages and scripting languages.

It's much easier for C to be backward compatible, not because it's a system language per se but rather because it's so minimalist. You don't have problems dealing with unicode when you don't deal with unicode. You don't have a problem dealing with ranges and iterators when your most advanced looping construct is for(i = 0; i Portability issues in C are mostly between architectures, not standard versions. The fact tha…

By the way, recently I realized that what I want is a language that does less instead of doing more. After all all these niceties with ranges and iterators are just syntactic sugar. Write pythonic code and face problems when porting this algorithm to C because C has nothing of Python niceties, which stop looking so nice at this point. Write non-pythonic C-like code without classes but with `__slots__` and functions all the way down and get not only the code that can be ported much easier to C and many other languages, but, surprisingly, also the code that is much more accessible and easier to reason about.

Re: Sunsetting Python 2

#252

Post by jedimastert »

sweettea wrote: post #142

Sad to see them forging ahead on their plans to drop Python 2, and it would have been nice of them to point out that other volunteers will provide security fixes for Python 2 compatible runtimes. I've switched to Tauthon [1] for over a year now and have been quite pleased. Consider the switch yourself rather than rewriting your code. [1] https://github.com/naftaliharris/tauthon

In your opinion, is there any particular reason to start a new project, or is this disappointment for lack of legacy support?

Re: Sunsetting Python 2

#253

Post by DangerousPie »

drej wrote: post #159

What surprises me about this is that the documentation for Python 2 does not explicitly say that the language version is about to be unsupported, see e.g. https://docs.python.org/2/library/zipfile.html Contrast this with the Postgres website, which tells me I'm browsing old docs (because Google still offers old links), see e.g. https://www.postgresql.org/docs/9.2/tutorial-window.html Are there plans to add a banner t…

100% agree. I think every Python 2 doc page should have a big banner on top mentioning the EOL status of Python 2 and linking to the coresponding Python 3 page (if there is one). Googling "python + keyword" far too often still takes you to the Python 2 docs.

Re: Sunsetting Python 2

#254

Post by Scarblac »

EnderMB wrote: post #146
Scarblac wrote: post #105

Earlier quoted context omitted.

Well, the largest companies in the world don't have to rely on volunteers, they can pay for continued support.

I've never worked for a company of that scale, so I don't know how hard it'll be to find legacy developers, but I imagine it'll take at least several years to port a public-facing system to Python 3, if they were to start now. If they start in the next few years, I imagine it'll be much harder to find ready and willing Python 2 developers with the experience to navigate such a big code-base? They could hire consultan…

Well Python 3 came out in 2008, this date is known since 2014. If they start in the next few years, yes you'd expect it to be expensive.

But distribution vendors like Red Hat and Ubuntu have Python 2 in long term support distributions that they will give support for a few more years, so it is possible to use those. And after that they offer more years of support to paying customers. It's not necessary to hire programmers to do this for the companies needing support themselves.

Re: Sunsetting Python 2

#255

Post by cheez »

normalhuman wrote: post #235
jedberg wrote: post #62

In 2015, there was no way I could have moved to Python 3. There were too many libraries I depended on that hadn't ported yet. In 2019, I feel pretty confident about using Python 3, having used it exclusively for about 18 months now. For my personal use case at least, this timeline worked out well for me. Hopefully it works out for most everyone. I can't imagine they made this decision without at least some data backi…

> For my personal use case at least, this timeline worked out well for me. Hint for Hacker News commenters: 99% of the time, if your phrase includes the term "use case", you can safely just delete it and lose no meaning. It's just a way to say "for me" or "for them" that sounds more technical but really isn't. In this case, you could say "This timeline worked out well for me".

Are you sure you are indeed a normal human?

Re: Sunsetting Python 2

#256

Post by steveklabnik »

Stay_frostJebel wrote: post #151

Kill it with fire! Should have been removed years ago. It actually causes real issues, such as build systems that only support python2 but not python3. The sooner python2 dies, the better. I need to use both python2 and python3 since otherwise quite several software projects refuse to compile from source. An example is Mozilla's mozjs. http://www.linuxfromscratch.org/blfs/view/svn/general/js60.h... Also increases you…

> Also increases your trust in the company creating Rust when they are too lazy to fix their whole build suite.

Mozilla is not in charge of Rust in any way.

Re: Sunsetting Python 2

#257

Post by carlmr »

j-pb wrote: post #127
carlmr wrote: post #85

Earlier quoted context omitted.

Is ROS used in production for this? I've only seen it used for prototyping.

Yeah, Built Robotics[1] apparently does, at least according to their hiring page. And I think Audi is using it for their self driving cars... https://www.builtrobotics.com/

I mean hiring page doesn't mean they actually use it in the product. You can use ROS for designing the system, but in terms of performance it has issues if you try to use it in an embedded context. I still kind of doubt they have it running in their product.

Re: Sunsetting Python 2

#258

Post by CogitoCogito »

bigiain wrote: post #231
CogitoCogito wrote: post #43

Earlier quoted context omitted.

I think you're blowing it out of proportion. Call your project some other name. Say that your interpreter is compatible with Python 2 and that you provide continued support. It will probably be hard to find volunteers because most are moving/have moved to Python 3, but I don't think the threat of lawsuit will scare anyone away except those who want to infringe on PSF's trademarks and in that case...good?

I'm finding it difficult to understand why anybody thinks supporting Python2 should be done by volunteers? What's "in it" for the volunteer? How much "fun" does "Supporting an old language where the original developers have moved on to a newer and more interesting version of the language, but there's a bunch of complaining people who still want the old language to be supported but they aren't offering to pay for it"…

> I'm finding it difficult to understand why anybody thinks supporting Python2 should be done by volunteers?

I 100% agree. I was just responding to the parent post.

Re: Sunsetting Python 2

#259

Post by CJefferson »

avian wrote: post #51
CJefferson wrote: post #21

Earlier quoted context omitted.

The biggest problem is having a place to organise. The Python foundation is threatening to sue anyone continuing something called Python 2, or even a similar name. Finding volunteers who are willing to get sued to work on Python will be hard.

Are you referring to that "Python 2.8" thing [1] that happened some years ago? A comment from PSF [2] makes it clear that they don't have any issue with people continuing to develop their Python 2 codebase further. It is what their open source explicitly permits after all. They do have issue with people using a trademarked name and not making it clear that their work isn't endorsed by the Python Foundation. [1] https…

For me, if I want most existing python2 programs to work, I'd want to release my program as called "python", so "#!/usr/bin/env python" continued to work.

If I'm not allowed to do that, then you are already adding a lot of friction to continuing to support existing python installs.

Although, I wonder if you could claim you need to call it that for the programs to work?

Re: Sunsetting Python 2

#260

Post by ueiuestu »

jedberg wrote: post #62

In 2015, there was no way I could have moved to Python 3. There were too many libraries I depended on that hadn't ported yet. In 2019, I feel pretty confident about using Python 3, having used it exclusively for about 18 months now. For my personal use case at least, this timeline worked out well for me. Hopefully it works out for most everyone. I can't imagine they made this decision without at least some data backi…

This makes me wonder, are there any software frameworks and languages that are specifically built for multi decade use?

Where they only release security updates and no breaking changes?

Post reply on HN