On Sep 2, 2018, at 12:32 PM, Nick Coghlan ncoghlan@gmail.com wrote:
On Fri, 31 Aug 2018 at 10:52, Donald Stufft donald@stufft.io wrote:
I also know that Nick has a few concerns with accessibility to the wider community, which I believe I addressed in the other email thread, but to reiterate, I don’t think that the wider community cares one way or another, and I think the biggest benefit comes from being able to tailor this process to what we need.
I think the wider community cares more about "Is it in a PEP?" than you believe, as we had folks outright refusing to accept Description-Content-Type as legitimate when it was only defined in https://packaging.python.org/specifications/core-metadata/#description-conte..., with no PEP to back it up as being an officially endorsed interoperability standard.
Who refused to accept it? Is there any indication that they will only implement PEPs, or that they just wanted it defined as a standard before supporting it, and they don’t care how we define it as a standard.
Attempting to build legitimacy and credibility for a random collection of text files in a GitHub repo rather than continuing to use https://www.python.org/dev/peps/ feels like a pointless time-wasting distraction to me. It's not like PEP-level proposals come up all that often - even with the 5 year delay induced by the PEP 426 detour, PEP 566 still only had a handful of major changes in it (Description-Content-Type, Provides-Extra, the canonical conversion to a JSON-compatible data structure, allowing the Description to be in the metadata body, and explicitly referencing PEPs 440 and 508).
How much of “PEP-level proposals don’t come up that often” is because people don’t want to engage the PEP process? I know personally I’ve suggested more than one person write a PEP, and they balked at the idea. It’s entirely possible they’d also balk at this idea too, it’s impossible to know for sure, but for better or worse, the PEP process has a perception amongst some people as being a heavy weight bike shedding exercise. Some people also refuse to join distutils-sig to engage the PEP process, because of the history of toxicity of those mailing lists.
I know of *several* “PEP-level” changes I’ve got on my personal back burner that I’ve avoided because the current process aligns itself well with bike shedding and arguments. The PEP process itself encourages the behavior that burnt out Guido, because it uses archaic tools that incentivize people to repeat the same tired arguments over and over again. To comment on old versions of a proposal and straw man the hell out of them. I’m not exactly the thinnest skinned person who participates in the PEP process, and every time I think about writing a PEP, I feel like I have to mentally prepare myself. Some of that is the nature of “politics”, and no system will ever solve it. Some of it is due to the process and tools we use, and we can’t ever fix that for ourselves without breaking away from whatever python-dev does.