Off the top of my head, no. I’m just pointing out that there’s time to respond before things get shitty. This isn’t Redis rug-pulling their license. The community can be considered. Thoughtful.

This does suck, but there’s time. I doubt the actual legal hand over has happened yet, and OpenAI is (likely, I don’t actually know) an absolute mess, HR/structure-wise, internally. If the aquisitions I’ve been a part of in the past are any indication, there’s a 1-2 year ramp before things really start to go to shit within Astral (assuming there isn’t an immediate massive exodous of talent). Or I’m totally wrong!

I was hoping to hear “I am moving away because the JVM sucks to administer”…oh well. A man can dream.

Java 8 was a thing for a long time (source administered Hadoop clusters that were - and possibly still are - stuck on Java 8).

Java 8 was analogous to 1.8…for reasons.

I wanna say Java 11 (the version after 8) came out around 2011? After that the release cadence was somewhat steady. I think Java 21 landed around 2021?

(Note: I refuse to actually look any of this up.)

Edit: my refusal to look anything up immediately bites as someone else pointed out:

1. There was a Java 10
2. It was released in 2018.
3. Both of these facts helped me remember Java 9 being released.

(Note: I continue to refuse to actually look anything up)

They keep the feature list fairly up to date:

https://asahilinux.org/docs/platform/feature-support/overview/

https://lap.dev/lapce

A potentially interesting alternative to Zed without the AI…

(That I have not used, so I don’t know if it’s good or bad)

People’s Front of Judea will triumph!

It wasn’t clear whether they have a connection proxy in front of the postgres instance. PosrgreSQL connections are expensive, so something like pg_bouncer could also make a big difference here.

(I realize the point was to test python web servers, but it would have been an interesting additional metric.)

Just some numbers with a Z next to them…

Turning them into Zombies is certainly a way of dealing with orphans, sure…

SIGKILL is a good way to orphan children. Or end a DB transaction without properly closing it (so the server socket stays open, or the transaction lock isn’t released…) It exists for a reason, but docs usually caution against it because it can cause unexpected behaviors.

https://devnull-as-a-service.com/

So you’re say remove emacs and replace it with vi? I agree!

Cowards are too afraid to place vi anywhere on these axes…

This is how someone cracked Okta a few years back: https://medium.com/@rajat29gupta/bcrypt-and-the-okta-incident-what-developers-need-to-know-9d13a446738a

I’m still skeptical about passkeys. Tying your auth to a specific physical device feels like a recipe for being locked out (as has happened to many people at work already…requiring me to remove their passkey so they can get back in to their account…)

Good point, but it’s “Custer”, not " Custard".

Although I kinda like the idea of a trembling, gelatious shape being the asshole that led the charge at Little Bighorn…

Also on Fedora.

Discover no longer shows a progress bar in the lower right for updates from rpms. Flatpaks do work, though.

Opening Spectacle immediately opens the area select screen, but causes any open menus to close, negating the usefulness of screenshots somewhat…

(Also, a recent update made my boot take way longer. Like…3 seconds to 1-2 minutes. It’s weird, and seems to be some kind of hardware issue…)

https://asahilinux.org/fedora/#device-support

I think the best answer is “support is progressing”.

I only know gnomes from this documentary from the 80s. Perhaps it was propaganda from Big Gnome, but they seemed nice…