@mcc Thank you for spending the time needed to engage with this (and for explaining it along the way).

🎤 PyCon AU is calling all Pythonistas to share their insights, stories, and innovations at our main conference AND exciting specialist tracks:

🧠 Data & AI
🔬 Scientific Python
📚 Education

Whether you're a seasoned expert or a passionate newcomer — we want to hear from you!

⏰ Submissions close soon pycon.org.au/program

#PyConAU #Python #Data #Education #AI #Science

@mhoye It's also one of those weird things where people will sometimes hold themselves to *higher* support standards than larger projects or even commercial vendors. Asking them if they have considered just not doing that (at least, not for free) can go a bunch of different directions (usually good directions, occasionally unhelpful ones).

Yo, Trans folk, little wisdom from my Black family…

You ain’t gone win, babe. Don’t try to win cause the game is rigged and ain’t no way you gone win. Your job ain’t ta win, your job is to make the work a bit better for those who come after. And you ain’t gone do that crawlin into no hole.

Be out, be seen, be seen happy, and make it home alive. You ain’t gotta start no fight, you ain’t gotta win the war.

You gotta let that little kid see you, so they know their fire is worth burning for.

Always forward babe. Never back.

@brettcannon Ah, wait, I forgot you were also an author on the build system PEP *and* the static core metadata PEP. Absolutely no potential for expectation warping in those, nope, none whatsoever :)

@nedbat I generally prefer "actual == expected" (for no specific reason beyond it feeling closer to what I want to express), but I settle for "whichever order gives the most useful diff for complex comparisons" (I find diffs that show how turn the actual result back into the expected result are much harder to interpret than those that use the expected result as the baseline and show how the actual result differs)

On the PSF Blog: A summary of the "An Uncontentious Talk about Contention" by Mark Shannon during the #Python Language Summit at #PyConUS 2025
https://pyfound.blogspot.com/2025/06/python-language-summit-2025-uncontentious-talk-about-contention.html

@cfbolz I have now located the volume controls in the printer's settings and told it not to make any more noises :)

@cfbolz The sound was at least soft enough that we knew it wasn't any of the smoke alarms. It eventually occurred to me that "new sound" correlated with "most recently added device" (network printer replacement). I forgot I had set the new one to poll for firmware updates, and it seems it beeps if it finds an update and hasn't received approval to install it yet.

@brettcannon That was one of the most complex PEPs since version numbering was standardised, so yeah, a little warping of expectations seems inevitable :)

Some device in the house is intermittently beeping. I do not know which device. We have many devices which it could be. It is infrequent enough that standing near them waiting for the next beep is impractical.

@brettcannon Well, you got me all curious now (and if it's the thread I think you meant, it did prompt a couple of new questions/ideas. Nothing major, though)

@mcc Interesting that the centralised services do offer revocable terms, though, since the client caching problem would affect even them. So yeah, your concerns are very well formulated, and my apologies for replying before reading more than the issue title.

@mcc We went through this with PyPI several years ago. The problem is that the internet never forgets: even when the service provider deletes uploaded content on request, they *can't* guarantee that all copies of publicly (or even selectively) available content are gone, as most copies are outside the service provider's control. The legal framing of that problem generally shows up as an irrevocable license for verbatim distribution instead of as obliging consumers to respect deletion requests.

@glyph Reconciling (or at least learning to mostly live with) the fact that even the vendors that seem to be exceptions to the rule of predatory capitalism often aren't is a big part of what led to me taking a few years off from particularly active community participation. And I *started* with a decade in the defence industry, so my expectations for corporate behaviour weren't high to begin with - I merely hoped to see a bit more *long-term* self-interest over actively destructive fad following.

This (obviously, I hope?) does not apply to anyone who actually *does* know what EEG data normally looks like and how one would conduct a study like this. If you are a neuroscientist or a research psychologist who was not involved with the paper, I am SUPER interested in hearing what you think about it, regardless of whether you think it's amazing or terrible. Please share thoughts! Getting multiple *informed* perspectives on this will be very important.

@brettcannon Did you end up trying out https://github.com/beeware/podium?

@brianokken @hugovk That's phase 3 and requires an ecosystem level readiness assessment, rather than just a CPython level one (the GIL really does mask a lot of potential multithreading race conditions)

@webology When there are political factions pushing hatred for anyone that sees the traditionally rigid definitions as the social fiction they are, gender unfortunately *becomes* political even if it shouldn't be.

@bitprophet Eh, I can kinda see it. It's certainly the *complement* of nothing, and "opposite" is a bit ambiguous where spectra are concerned.