Live data from Hacker News

Debian 12 “Bookworm”

debian.org

231–240 of 241 posts

Re: Debian 12 “Bookworm”

#231
post #228
post #224

Earlier quoted context omitted.

Proprietary extensions need to live in the opcode space reserved for custom extensions, where it'll overlap with other vendors' own custom extensions. If they conflict with RISC-V's own space, the chip can't use RISC-V trademarks. And, of course, the RISC-V Consortium will never adopt proprietary extensions, as that goes against its core values.

What matters is what is on the market, and how it is made available. Everything else matters as much as FSF point of view on something being free or not.

That's why there are licenses and agreements.

So that we do not have to rely on anybody's goodwill.

Re: Debian 12 “Bookworm”

#232
post #231
post #228

Earlier quoted context omitted.

What matters is what is on the market, and how it is made available. Everything else matters as much as FSF point of view on something being free or not.

That's why there are licenses and agreements. So that we do not have to rely on anybody's goodwill.

And since RISC-V allows for extensions, all bets are off regarding successful deployments being extension free.

Just like many open core products that really need the full deal for being usable in a proper way.

Expecting otherwise is whishful thinking, or academia/maker community focus.

Re: Debian 12 “Bookworm”

#233

Earlier quoted context omitted.

I want to downvote you for sentimental reasons. I won't but I want to.

Do it, and tell the world how I mistreated all kinds of interesting (but small) projects, if it makes you feel better. :-D Here a list of almost all the OS distris I've left out: https://distrowatch.com/search.php?ostype=All&category=All&o... Besides mentioning Slackware for historical reasons I should have also mentioned NixOS, which I forgot. The former was the first Linux distri, the later gained some respectable…

why is using alpine for containers a recipe for trouble?

Re: Debian 12 “Bookworm”

#234
post #232
post #231

Earlier quoted context omitted.

That's why there are licenses and agreements. So that we do not have to rely on anybody's goodwill.

And since RISC-V allows for extensions, all bets are off regarding successful deployments being extension free. Just like many open core products that really need the full deal for being usable in a proper way. Expecting otherwise is whishful thinking, or academia/maker community focus.

There already are. As long as they're in custom space, there's no issue.

Re: Debian 12 “Bookworm”

#235
post #226

I'm not a Linux expert, I just want an OS that works well, and that's debian for me. Now, how do I upgrade? Just apt upgrade? Does it matter that I have i3wm instead of whatever the default is?

The upgrade instructions are in the release notes. It looks long but is pretty straightforward if you haven't installed software from outside the debian ecosystem or created a frankenstein mix of packages from non-stable, non-backports distro versions. https://www.debian.org/releases/stable/i386/release-notes/ch...

I've installed vs code by downloading a .Deb, not an apt repo. Does that count as "software Frankenstein"? Follow up question, what's the worse that can happen? Vscode will break, or might the os get corrupted?

Re: Debian 12 “Bookworm”

#236

Earlier quoted context omitted.

Do it, and tell the world how I mistreated all kinds of interesting (but small) projects, if it makes you feel better. :-D Here a list of almost all the OS distris I've left out: https://distrowatch.com/search.php?ostype=All&category=All&o... Besides mentioning Slackware for historical reasons I should have also mentioned NixOS, which I forgot. The former was the first Linux distri, the later gained some respectable…

why is using alpine for containers a recipe for trouble?

TL;DR: It's slow and has all kinds of issues. If all you want is small containers use Distroless.

Musl is still a kind of experiment. I would not recommend running experiments in production…

Just a few random links:

https://nickjanetakis.com/blog/benchmarking-debian-vs-alpine...

https://news.ycombinator.com/item?id=28312433

https://www.linkedin.com/pulse/musl-libc-alpines-greatest-we...

https://martinheinz.dev/blog/92

https://www.linuxquestions.org/questions/linux-software-2/mu...

https://unix.stackexchange.com/questions/729342/performance-...

https://github.com/rust-lang/rust/issues/70108

https://news.ycombinator.com/item?id=23080290

https://vector.dev/highlights/2020-07-09-add-musl-and-glibc-...

There are more. Much more of those!

Musl exist in large parts just for ideological reasons. It still rides some hype in industry as industry hates GPL software…

Re: Debian 12 “Bookworm”

#237
post #44

One thing I really appreciate about Debian is that when a new stable release comes around, I can just upgrade and be reasonably sure nothing bad will happen. It's not exciting, but a fair amount of the time, this is what people expect from their operating system. Support my hardware, give me the software I need, and stay out of my way otherwise. And that is what Debian does very well.

I heard when Bullseye came out that I should wait a bit as the initial bugs were found. I'm wondering if that was true then or now.

It depends. I use Debian on a few machines at home, if something were to break, it wouldn't be a huge deal.

If I had dozens of desktops and/or servers to take care, I would probably take the time and upgrade a few non-critical machines to see how it goes, provided I have the resources for that.

If in doubt, there's nothing wrong with waiting a month or so to see if others run into trouble. It's what I did when I worked as a Windows admin, and it saved me from major headaches more than once. (Admittedly, updates causing trouble is more common on Windows than on Linux in general and Debian specifically.)

Re: Debian 12 “Bookworm”

#238

Earlier quoted context omitted.

why is using alpine for containers a recipe for trouble?

TL;DR: It's slow and has all kinds of issues. If all you want is small containers use Distroless. Musl is still a kind of experiment. I would not recommend running experiments in production… Just a few random links: https://nickjanetakis.com/blog/benchmarking-debian-vs-alpine... https://news.ycombinator.com/item?id=28312433 https://www.linkedin.com/pulse/musl-libc-alpines-greatest-we... https://martinheinz.dev/blog/9…

Thanks for the links. I'm just an hobbyist, but Alpine is often marketed as being great for containers. Interesting to see another perspective

Re: Debian 12 “Bookworm”

#239
post #209
post #112

Earlier quoted context omitted.

While I'm glad to see proprietary firmware both included and segregated in its own repo, I'm wondering why RISC-V wasn't added as a supported architecture in Debian 12? It seems like that supporting at least an open ISA would move closer to the possibility of what Debian wants to see happen... so why isn't it as a distro helping to make it happen?

At the end of the day, this was because the hardware ecosystem (and the hardware support in Linux/bootloaders/etc) isn't really "ready" yet, and also there was slow and unclear communications between the porting team, the core Debian teams and one of the hosting providers. Eventually the teams provisionally approved the port, but after that the needed actions weren't done in time for the archive-wide rebuild that hap…

Sorry for the lag in my reply (I just noticed your response)... I very much appreciate the info as I found the lack of official RISC-V support in 12 surprising. I hadn't read about the blockers to being ready for the release which makes perfect sense. Looking forward to seeing in in 13!

Re: Debian 12 “Bookworm”

#240
post #225
post #70

Earlier quoted context omitted.

yes, as the script adds/replaces the sources.list lines, then apt i downloads the packages, only then it starts the installation of them

But you can't (easily) get the deb package for systemd-resolved after the upgrade has (possibly) broken DNS?

yeah, that's not ideal, or possibly very bad UX, depending on how the update happens. (if the running stub resolver of resolved is not shut down, or at least an effort is made to fix resolve.conf during removal of the old package, then it might cover most of the users. as far as I understand the 11 -> 12 upgrade works without manual intervention, and this is an edge case, as systemd-resolved is not enabled by default on Debian 11.)
Post reply on HN