Comment by forgottenacc56 | Hacker News Reader

Comment by forgottenacc56 | Hacker News Reader

Parent Full thread

There's nothing that will convince the Python 3 haters. So silly. I use nothing but Python 3 and I rarely if ever find I can't move ahead for lack of a library. The Python2 only libraries are just being replaced with more up to date alternatives. This article is total FUD and garbage from yet another Python 2 ideological holdout dogmatist.

This article would have you believe that if you tried to use Python 3 that within hours you would hit a library roadblock that would be a showstopper. Completely wrong, and you have to ask why the author wants to create this perception.

Python 3 works beautifully and is a pleasure to program.

If you WANT to hate Python 3 then you'll find no lack of ways to hate on it for the next 20 years (HEY! I found some package built by a fellow Python2 holdout! See, it simply is not possible to write the code to upgrade it! PYTHON 3 IS FINISHED! THE SKY IS FALLING!) etc etc ...

If you want to succeed with Python 3 then that's easily done, you just get on with it.

No one using Python 3? Flat out lie. Prove me wrong by scientifically explaining your assertion that no one uses Python 3. Lies, spoken to back the wishes of the Python2 holdouts to sabotage Python 3 - they hate Python 3 and will say any lie to bring it down and spread the FUD. I'm wrong? Prove it in detail with science. Alternatively make vague, handwavey statements about downloads, as though that defines how many people are writing code today with a language.

Which article did you read - I don't see any of the claims you are suggesting they made. I even did a search for the word, "library" - it never appears.

If anything, based on my read of the article (twice, I had to read it a second time after reading your comment) - I think it's actually written by a Perl advocate who is ruminating on the future of Perl 6, and just using Python as an example of a team that shipped code fast and limited the amount of stuff they broke.

Are you sure we're reading the same article?

Did you read a different article? The author only had positive things to say about python3 except its adoption (which has been noted to be an issue multiple times before this article existed). You are blowing the author's intent way out of proportion.

I've found exactly the same thing.

As someone writing applications using Python 3, I found the switch to be easy and I haven't had any problems with libraries. I've switched from some of the old ones I was using to more modern alternatives that are being actively developed.

It annoys me that there are people on Python 2 that haven't really tried to make the switch and stand on the sideline shouting insults at Python 3. It's a tremendously self destructive thing to do for all involved.

To anyone looking to get started with Python, there's no real reason to start with Python 2 these days.

Libraries are a problem - it's difficult to switch unless all your dependencies are Python 3 compatible. The only ways to fix it are to either use an alternative, or perhaps to fork or contribute patches to the project - both of which take up valuable time when your code is still functional under 2.7.

I'm currently creating 2.7/3.3+ compatible code in whatever I create, but I'm still deploying to 2.7 hosts. For most of my projects this is due to just one package which has some Python 3 related bugs, so hopefully I can just flick the switch soon and use 3.4.

It's only easy when your project depends on a few popular packages (eg. the top 200 where 86% are currently Python 3 compatible [1]), it's much more difficult if you've got a long list of dependencies.

[1] https://python3wos.appspot.com/

Yep, for me, distributions are the biggest thing holding things back right now, because the existing ones are out there, and will be deployed for a long time.

If deploying a .com website, you can control your dependencies completely, and it's much easier than if you are deploying a system that must interact with existing systems the user/customer might have (for instance, management or monitoring apps).

Thankfully that's not a huuuge class of systems, but it's what I typically work with :)

I suspect most people don't have to deal with that issue (fortunately, because that must suck). Is there not some way of wrapping the entire binary with your distribution? If I was working in that space I think I'd probably choose a different language, if possible.

As you say, it's not a problem for most of us as we get to dictate the platform.

If you're wondering how you can do above / why it's not such a big deal: Python 2.7 has everything you need to write forward compatible code. It gives you all the __future__ dunders, the only extra thing you need is a compat module [1]:

See kaleidoscope of examples:

- https://www.google.com/search?q=site%3Agithub.com+compat.py+...

- https://github.com/search?utf8=✓&q=extension%3Apy+"if+PY2"&t...

[1]: http://lucumr.pocoo.org/2013/5/21/porting-to-python-3-redux/

I'll get downvotes for this, but my reason for sticking with python2 isn't library support, it's that I prefer python2's strings. Sure, it made unicode handling less painful in cases where encodings are always known and everything should be in unicode, but it makes things much harder for people working with bytes interfaces that want to treat all non-ascii bytes agnostically.

But Python 2 strings are bytes. So in Python 3 just use bytes everywhere. Or am I misunderstanding something?

Some current facts: ~17% of packages on PyPI support Python 3 and ~5% of packages downloaded from PyPI are downloaded for Python 3.

If you start a project today then yes, start with Python 3. But if you have a Python 2 codebase you're likely to have (1) one or more dependencies that haven't been ported yet, (2) no solid economical reasons to convince your employer to switch or (3) no sufficient technical reasons in switching at this point (which may change as more Python 3.x versions are released).

You're referring to packages that most people will never ever use, or have heard of.

What you should be referring to is http://py3readiness.org/ — which shows the opposite picture. Above 80% of the most popular packages are available on Python 3.

>~17% of packages on PyPI support Python 3

And how is that determined? Setuptools/distutils classifiers? Then is completely meaningless, most don't even bother including those.

And a big chunk of the packages will never be Python 3 compatible simply because they're abandoned and shouldn't be counted either in my opinion.

I was scared off of Python3 until recently for no good reason. When I started a new development effort I made the switch and have had almost no problems besides a few small configs I had to work through. Besides that and going over some new syntax and recommended techniques I spent maybe 1-2 days all in. It's seriously not that big of a deal.

The problem isn't that there are problems with going to Python 3. The problem is there isn't enough benefit to make any change worthwhile for most users.

Reply on news.ycombinator.com