[pytest-dev] A Short Personal Post Morten for the pytest exodus

[Ronny Pfannschmidt]

Hello everyone,


** This was originally intended to be send out mid April but got 
intentionally delayed to ensure Holger could do the communication for 
the resolution, i finally picked it up again to send it out **


After failing a number of times to sit down and actually throw together 
a reasonable post mortem, I decided to at least try to write up a basic 
summary.


Before continuing please take a look at Adrin Jalali: Open Source - CoC 
- Conflicts <https://adrin.info/open-source-coc-conflicts.html>

His writing gives an outside perspective on the issues at hand.


Overall I don't want to attribute malice to Daniel, however we have 
built up a sad story about failed communications and a failed process. 
Early in the process just a bit after the first CoC Report on Daniel, I 
had a phone call with him and exited it under the impression that things 
are better now, the next morning I woke up to a complete “annihilation” 
of all his issues/prs (rage closed?).


At that point in time I reached the frustrating conclusion that I 
couldn't get through to him, and as far as I know we never got a 
satisfying explanation for that. That event also changed my perception 
of him as my trust was gone. In the weeks and months that followed the 
behaviour stayed the same and Bruno and I tried to get the CoC to act, 
as small little details that border on CoC violations and just drain our 
will to work on things kept accumulating.


At the same time I also didn't want to push for intense action, as I 
consider myself a “affected party” and would like to see someone with a 
more clear head on things cut out some clear decisions.


What I perceived then was the helpless situation that

 1.

    I couldn’t give Daniel the direly needed feedback in a sensible manner

 2.

    The CoC Team also didn't.


So the Situation kept “giving” - to the point where I would sometimes 
fail to communicate to Daniel the way I would like to as the frustration 
kept getting to me.I then quit the CoC team in frustration, but only 
with a relatively short unhappy note.

Some more weeks passed in which Bruno and I worked with each other to 
have a little outlet.


However when Bruno finally called quits, that setup was gone, I 
hesitated a while and then decided to join the exodus.


As far as i can tell in retrospect, there are many points where i see 
potential “what if’s”,

However when things happened I didn't see those.


My own inexperience at CoC enforcement plays a role, given the current 
situation I sure wish I had acted earlier and more. Also I tried to be 
nice even though the situation begged for a more direct and honest 
approach, after all bad behaviour needs early and effective feedback to 
correct and to be clear for all sides.


With my departure from the CoC team I felt that I shed myself of that 
part of the responsibility (as I didnt perceive an effect) , but it 
certainly is my responsibility to protect myself from bad behavior.


While I would prefer that this was done in an effective and positive 
way, I recognize that shortcomings in the communication between Daniel 
and me, The CoC team and me as well as my inexperience with situations 
of that kind, have contributed towards the current outcome.


After I sent out the short message about quitting as well, I underwent a 
number of emotions, but in the end I do feel a great relief, I have shed 
myself of a role that became negative to me.

Now I have quite some more emotions to unpack and reorient myself.


I want to be part of a CoC-Team again - but next time I want to actually 
be trained in the process,I do believe that CoC Teams are an important 
tool to keep communities working,but as Adrin wrote, one has to hone a 
particular set of skills to do it decisively and quickly.

I believe part of why the CoC process with pytest went so frustrating in 
the case at hand is that all 4 of the initial members are trying to be 
nice people, and I'm very sure the other 3 are way nicer than I am. 
However that particular trait hinders a certain positive harshness and 
in a way paves a road to hell with good intentions. It’s my impression 
that we all missed our opportunities to sort this out more favourable 
for everyone involved.


I also want to continue to contribute to the python testing ecosystem in 
one way or another,After all I had myself lined up for  a multi-year 
process of refactorings and minor breaking changes of pytest internals 
to get node structures, fixtures and test execution in general ready for 
the things to come.


What I see myself doing in the near future is something like minimal 
modular composable testing components.


I also can see myself returning to pytest, but that's not yet something 
for which I have a concrete idea.


For now I want to reinvent myself a bit and see where that leads to.


-- Ronny

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pytest-dev/attachments/20200520/9a323703/attachment.html>