@mgorny@social.treehouse.systems#GitHub is going great.
Remember the issue where PRs wouldn't show up on repositories? Well, it turns out it wasn't fixed at all. #Microsoft just "resolved" the incident and is rejecting all reports about it now.
https://github.com/conda-forge/amplitude-analytics-feedstock/pulls is one example when you can't see the PRs.
See e.g. https://news.ycombinator.com/item?id=48219231
@JSkier@social.linux.pizzaStill running, on an elliptical 😊 It’s been an adventure adapting to new disabilities. Still registered for the half marathon but I don’t think I’ll be able to run it without walking or a wheelchair this fall.
One thing I’ve enjoyed lately is working with an Italian fella and his QZ Fitness app. I made an #esp32 bluetooth bridge to get better data points, and help get some PRs for impro…
@cyrevolt@mastodon.socialDrinks and checking PRs, woop 🥳
@cyrevolt@mastodon.socialLots of patches are pending for @… - currently 8 open PRs, with even more to come.
Who could help out reviewing them? 🥺
https://github.com/oreboot/oreboot/pulls
@peter_mcmahan@mas.toPushing back on the idea that GenAI is "just another tool" for coding.
This comes up all the time in discussions on its use in open source software. People will fall back on the argument that it's like linters or autocomplete, and all that matters is whether the code passes muster. But it's not. In addition to the ethical issues, GenAI is a fundamentally different kind of "coding tool" and needs to be considered differently.
A comment to that effect in…
@wraithe@mastodon.social
@ripienaar@devco.socialUsing my little fisk-ai project to triage PRs, simple wrapper appbuilder file around `gh` cli, AI has no general bash tool or direct access to `gh` tool.
Wrapper in gist, the wrapper becomes the tools. https://gist.github.com/ripienaar/0f0cd37217d5f6417426a41ad273076a
@fanf@mendeddrum.orgfrom my link log —
agitjo: manage Forgejo PRs with AGit-Flow in Emacs.
https://codeberg.org/halvin/agitjo
saved 2026-06-23 https://dotat.at/:/LHM1Y.h…
@knurd42@social.linux.pizzaOne of my pet peeves with #GitHub (but maybe it's just me):
I'd never write "knurd wants to go to the cinema from home" as "go from home to the cinema" is way easier to understand.
But for some reasons github uses the odd order when it comes to PRs:
"knurd wants to merge 1 commit into kernelci:main from knurd:gitlab-workitems" (as shown i…
@azonenberg@ioc.exchangeClosed out another three tickets against ngscopeclient/libscopehal. The v0.2 release is getting closer every day.
Still a bunch of open things to fix, need to make a pass through pending PRs, a bunch of testing, some thunderscope updates...
But I'm hoping to get at least a RC out in the next week or two.
@memeorandum@universeodon.comTODD BLANCHE MUST BE HELD ACCOUNTABLE FOR PATTERN OF SERIOUS ETHICAL MISCONDUCT (DDF)
https://www.democracydefendersfund.org/prs/06.22.26-pr
http://www.memeorandum.com/260623/p48#a260623p48
@hynek@mastodon.sociali'm so fucking sick of it
i've got ONE extremely simple and extremely explicit contribution rule/brown m&m test: do not delete the PR checklist
since i'm busy getting structlog out, I’ve let the attrs bug tracker a bit off the leash
go & count for yourself how many PRs opened in the past month followed that instruction.
(and of course, this is not about bureaucracy; they violate many of the important items on that list)
@jerome@jasette.facil.servicesInteresting perspective
https://nolanlawson.com/2026/05/25/using-ai-to-write-better-code-more-slowly/
@frankel@mastodon.topAnd I can feel it coming in the air tonight, oh Lord 🎵🎶
Well, I've been waiting for this moment for all my life, oh Lord 🎶🎶🎵🎶🎵🎶
https://blog.frankel.ch/java-geek-weekly/133/
@azonenberg@ioc.exchangeI'm wondering how plausible it would be for the new ngscopeclient CI to have a private web form (probably just restricted by simple https basic auth with creds shared among project maintainers) to request a build of a specific commit in order to enable testing of PRs.
The idea is that this will still allow pre-testing of PR content before we merge them, but won't allow any possibility for an outside attacker to trigger a build on our infrastructure without a maintainer at least…
@mgorny@social.treehouse.systemsSince my PRs are taking a while to merge, if someone wanted to play #CrispyStrife with improved gamepad controls (one where selecting a dialog option doesn't attack the NPC):
#DoomCoding.
#games #Strife
@mgorny@social.treehouse.systemsCongratulations. Now I'm actively worried to open a PR fixing my issue, because there's already been a few #slop PRs opened attempting to fix it. And I'm literally worried that mine would be taken as another slop PR…
#NoAI #NoLLM #FreeSoftware #GitHub
