[spenrose]

AFAICT, this article repeats the disaster of PEP572: it makes almost no effort to discuss how Python is used. Specifically, it implicitly holds that the BDFL + core contributors model that served Python so beautifully from, say, 1992 to 2002 when Python had perhaps 2 orders of magnitude fewer users than it does now and ran on—how can we know—3? 4? orders of magnitude fewer machines, should be taken for granted as the default governance mechanism for Python in 2018, when it is a key part of the world's infrastructure. I have no idea how to manage something as important as Python, but reading python-dev and seeing the brilliant Tim Peters, who was indispensable to Python's early growth, argue for a significant change in the language's semantics by using it to refactor a few samples from his personal code base, strikes me as prima facie evidence that neither do the core contributors. Millions. Of. People. Write. Python. That. Affects. The. Lives. Of. Billions. Of. People. Start there.

[whyever]

> Millions. Of. People. Write. Python. That. Affects. The. Lives. Of. Billions. Of. People. Start there.

I don't think this is how open source development works. Probably most of the core contributors are doing it for pleasure, not for satisfying the needs of the users.

[spenrose]

Thanks, that's a much improved way of expressing my point.

[newen]

This idea that are large and influential project like Python is just the same as a hobby project I put on github and do in my spare time is ridculous. I am pretty sure most of the core contributors of Python are doing it as part of their 9-5 job.

Then there is the implication that because they are doing it for pleasure, they don't owe anything to anyone. It's ridculous. This software affects millions of people and it's not a dumb hobby project.

[xj9]

actually, by using Python you explicitly agree that PSF doesn't owe you anything :)

https://github.com/python/cpython/blob/master/LICENSE

4. PSF is making Python available to Licensee on an "AS IS" basis. PSF MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR IMPLIED. BY WAY OF EXAMPLE, BUT NOT LIMITATION, PSF MAKES NO AND DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF PYTHON WILL NOT INFRINGE ANY THIRD PARTY RIGHTS.

5. PSF SHALL NOT BE LIABLE TO LICENSEE OR ANY OTHER USERS OF PYTHON FOR ANY INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES OR LOSS AS A RESULT OF MODIFYING, DISTRIBUTING, OR OTHERWISE USING PYTHON, OR ANY DERIVATIVE THEREOF, EVEN IF ADVISED OF THE POSSIBILITY THEREOF.

[newen]

I'm talking about social responsibilities. Not about legal liabilities.

If there is not implied social responsibility that PSF keeps python running, then no one will be using it.

[megaman22]

Why? As long as python.exe runs code, people will use it. And there's too much money invested in it for people that have done so to let it just die.

[newen]

Because PSF says they will support Python? What happens when there is a bug? You fork Python? PSF gets 3 million in revenue per year. Where do you think the money goes? Python is not a hobby project.

[msla]

> Then there is the implication that because they are doing it for pleasure, they don't owe anything to anyone. It's ridculous. This software affects millions of people and it's not a dumb hobby project.

... no more ridiculous than the idea that software which millions of people depend on should be a commercial product, or a government project, and beholden to the politics of CEOs and/or government bureaucrats.

[whyever]

If it is part of their day job, the responsibility is towards their employer, not their users.

The millions of users are usually providing nothing in return for the software, I don't see how that entitles them to anything. In fact, I think it is the user's responsibility to make sure the software is appropriate for their use.

[s73v3r_]

That works for the rinky dink little file browser you put up on Github. It does not work for a project of this scale. Any core contributor who is unwilling to consider that point and treat this like a professional needs to step down, now.

[skybrian]

Wow, that's entitled. Did you know that Python has a money-back guarantee?

[s73v3r_]

No. For something of this nature, it is not entitlement to expect the people working on the core team to treat it like a professional. The entitlement comes when you believe that you can take a project like this, that millions of people are using to affect billions more, and treat it like a little hobby project.

[msla]

You're using a lot of words to say nothing concrete.

What, precisely, do you mean by "professional", and how is the Python team not being "professional" right now? Because the definition of "professional" I use is "Doing something for money, as a profession", with no implications about the quality of the work, and that's obviously not what you mean by it... right?

[skybrian]

Open source code is a gift. You don't have any right to demand this of them, and neither does anyone else. (What could possibly give you this right?)

If the governance of an open source project is bad enough, the way to fix it is to get a group of people together to maintain a fork. (Companies that depend on open source software can and will maintain their own branches.)

[s73v3r_]

"Open source code is a gift."

And the ability to work on it is a gift, too. If you're not going to take working on something as high profile as Python seriously, if you're just going to treat it as a hobby, then do everyone a favor and don't bother.

"You don't have any right to demand this of them"

As a user of Python, I absolutely do.

[munificent]

> Millions. Of. People. Write. Python. That. Affects. The. Lives. Of. Billions. Of. People. Start there.

How do you take that assertion and turn it into something actionable?

What, concretely, would they be doing differently if they fully signed on to the above?

[toyg]

I think what Raymond Hettinger said is a good starting point: slow down and allow the ecosystem at large to catch up. Make big changes in a slower manner, and if there is no widespread agreement on a PEP for more than a year or two, just drop it.

[spenrose]

1. I addressed this point by saying "I have no idea." I think the world has no idea. IEEE and other professional standards bodies are the only model I can think of, but it doesn't feel like a great fit. I tried to move the conversation forward. Maybe instead of challenging me, you could turn it into something actionable (or otherwise move the conversation forward). 2. Your "assertion" smacks a little of gaslighting. Python lies near the heart of the major Linux distributions that comprise, for example, AWS and Google Cloud. And they run services that directly or directly touch pretty much everyone in the industrialized world. Netflix by itself has 125M users, for example.

[TheDong]

You claim that the python core team should be aware python is used by so many people, but you have no clue what them being deeply aware of that looks like, so how can you possibly claim that they're not doing that now, that doing so will make any difference, and that it's "forwards" for them to internalize that?

I also am absolutely floored that you'd pull out the term gaslighting for the parent comment which makes no assertions and only asks you to elaborate on your point more.

I think you need to step back, re-read this comment chain, and go look at a mirror.

[philosopherlawr]

Learn from Javascript. As a full stack dev who came to python only recently, I think Python could really learn from what's happening in the js ecosystem.

[toyg]

To make sure it does the opposite of anything the JS ecosystem ever did? In that case, I agree.

[redsymbol]

Don't know why you're being downvoted. As a Pythonista and author of a Python book, I'm really curious. Would you mind suggesting a few aspects of Javascript's ecosystem to look at?

Thanks in advance.

[bluecalm]

It wasn't a disaster. Some people didn't like the final choice. Some liked. A useful feature was added, this as far from disaster as it can get.