Live data from Hacker News

Chimera Linux

chimera-linux.org

171–180 of 224 posts

Re: Chimera Linux

#171
post #118

Earlier quoted context omitted.

Oh sure, I still remember getting my first 1/4” tape of the GNU tools source code through the post. This is why it’s so hilarious to me when the kids nowadays complain that MacOS has different command line tools to Linux. “Why can’t they just use GNU?”. Get off my lawn! ;)

Ironically, BSD was already available when GNU was created. Since that is what macOS uses, I guess the question could be “why didn’t everybody just use BSD”?

I suspect two of the many reasons for GNU usage were

(1) crazy command options (it only takes one or two to ease your workflow over the what the system distro supplied, plus the same behavior for the same flags across platforms)

(2) the lawsuit (probably an understatement - it was Serious back then)

Edit: added "usage" for clarity, s/statment/statement

Re: Chimera Linux

#172
post #137

Even though I do not fully agree, I think this project has one of the most mature, nuanced, and balanced takes on systemd I have seen yet [0]. [0]: https://chimera-linux.org/docs/faq#what-is-the-projects-take...

This is also the stand of Void Linux by the way. Void is not systemd-free for the sake of it but to offer a Linux system that keeps its Unix heritage of small, loosely coupled components, is reasonably POSIX and portable (xbps), and lets the user pick his libc.

The Chimera founder was a Void maintainer. There is certainly a fair bit of philosophical alignment.

If Chimera is able to become as popular as Void is now, I would be happy. It would be amazing if it could be the next Arch. I am not sure what q66 wants for the distro but keeping the core distribution pure and having some successful derivatives would be cool ( if they kept to much the same vision ). But I guess I should let the project get out of Alpha first. :)

Re: Chimera Linux

#173
post #109

Earlier quoted context omitted.

GNU glibc has some gratuitous extensions over ANSI C/C99. Unix utilities have long non-standard and non-POSIX based flags.

So what? The OP wrote: > Those hacks that allow for the ease of use in a GNU userland make programming a massive pain in the ass where nothing "just works" the way it should, because GNU programs and libraries are in flagrant violation of every specification and rule of thumb that has ever really mattered for Unices. Just because the tools offer additional functionality does not mean they "violate every specification…

Posix just states ed, but a pure Unix doesn't have a "default" editor.

But vi/nvi it's useful because you can use commands to replace text inline, or pipe text parts to commands.

Re: Chimera Linux

#174
post #163

Earlier quoted context omitted.

Yeah pretty interesting. Personally I never liked systemd because I just didn't see the need for it and I don't like changes forced on me. I'm not a team player and I hate being told to 'just go with the flow'. The whole idea of FOSS for me is to be in control of my computer. Alnd I just don't like the character of Poettering and yes that matters. For a much more extreme example look at ReiserFS. It was totally shitc…

I’ll never understand what that guy did that got all of you in such a frenzy. Because Stallman or Torvalds are such socially apt and charismatic people, or what? Yet all those systemd opponents out there don’t seem to have a particular problem with them. Poettering might have opinions on what Linux should be that don’t overlap with yours, but systemd solves actual problems and works better than any init script shenan…

Well, both Linus and RMS have attracted their share of controversy. In the case of Linus, while you may not always agree with him, there seems to be broader support for his engineering choices. He has multiple projects that have dominated their spaces on their own merits. Whatever you think of Stallman, there is no denying his historical contribution.

For myself, I am more bothered by RMS. Although I know Chimera has shunned GNU for technical reasons, I would personally prefer a world where all GNU software has been replaced by alternatives. In my view, we would be better off to be free of RMS and the politics he has created around around “GNU / Linux”. The GPL itself is an important option in Open Source and, while it is not my favourite, I wish it and the projects that adopt it much success. However, the GNU Project itself is becoming an impediment ( my views only of course ). Luckily, the GNU project represents quite a small list of software projects despite the way people talk about it. Replacing GCC, Glibc, and Coreutils would be enough. Of these, the most damaging and the hardest to replace is probably Glibc.

Anyway, my intent is not to trigger people with such opinions in this thread. I do not want them to reflect badly on Chimera. My only reason for chiming in is to illustrate that Poettering is not unique in collecting detractors.

Finally, systemd did solve real problems at the time. There are other solutions now though so let’s not render them inoperable. Also, not everything systemd wants to be now is better than the alternatives or even what came before. Not liking systemd does not mean wanting to go back to init scripts.

Btw, the system I use most daily uses systemd. I also use it in a project at work although, there, we have a strange bug “rarely” at shutdown that systemd is implicated in and I wish it ( and is ecosystem ) were easier to understand. So again, I find the Chimera approach attractive.

Re: Chimera Linux

#175
post #163

Earlier quoted context omitted.

Yeah pretty interesting. Personally I never liked systemd because I just didn't see the need for it and I don't like changes forced on me. I'm not a team player and I hate being told to 'just go with the flow'. The whole idea of FOSS for me is to be in control of my computer. Alnd I just don't like the character of Poettering and yes that matters. For a much more extreme example look at ReiserFS. It was totally shitc…

I’ll never understand what that guy did that got all of you in such a frenzy. Because Stallman or Torvalds are such socially apt and charismatic people, or what? Yet all those systemd opponents out there don’t seem to have a particular problem with them. Poettering might have opinions on what Linux should be that don’t overlap with yours, but systemd solves actual problems and works better than any init script shenan…

I'm not a fan of Stallman either. I think he focuses too much on the legalese around licences. Personally I don't care about licensing very much at all. I ignore all licenses and terms & conditions anyway (and to be perfectly honest; who actually reads them?)

Nor commercial involvement (e.g. RedHat/IBM or Ubuntu) for that matter. This is why I'm better off with BSD anyway because there's so much less commercial involvement.

And Linus yes he can be abrasive but usually I would agree with him more.

> I call your proficiency with Linux into question (don’t take this in spite, please, it’s in no way meant personally).

Well I'm certainly not very good with Linux anymore since I've used it less and less recently, gravitating more towards the BSDs. So that's not a bad point, I know how to start and stop systemd services and a bit on how to get logging back etc. But the point is, I never really wanted to learn something new. I've used FreeBSD since version 6 and it's hardly changed at all. And it works totally fine for me. But "being good with Linux" is not a goal for me anyway. I don't currently need it for my career anyway.

In the mean time I've been on Macs, but I really really abhor the way that Apple is constantly changing things around on me too (they are far worse at this than Linux) and they are consumerising the OS rapidly. So when I went back to FOSS I decided to go for BSD for this reason.

Re: Chimera Linux

#176
post #163

Earlier quoted context omitted.

I’ll never understand what that guy did that got all of you in such a frenzy. Because Stallman or Torvalds are such socially apt and charismatic people, or what? Yet all those systemd opponents out there don’t seem to have a particular problem with them. Poettering might have opinions on what Linux should be that don’t overlap with yours, but systemd solves actual problems and works better than any init script shenan…

Well, both Linus and RMS have attracted their share of controversy. In the case of Linus, while you may not always agree with him, there seems to be broader support for his engineering choices. He has multiple projects that have dominated their spaces on their own merits. Whatever you think of Stallman, there is no denying his historical contribution. For myself, I am more bothered by RMS. Although I know Chimera has…

I generally just don't want to see anybody being attacked for the work they do; I'm fine with people criticizing RMS, since that's generally because of his views that deserve to be condemned, but I wish to see no hostility towards maintainers of GNU projects or systemd

the GNU codebases may have their issues (but not always; there are some very high quality ones), but that's still not a reason for personal attacks, and frequently the maintainers are just trying to do their best and don't deserve any of that shit

ideally I'd like to see an ecosystem where all the projects are represented, where the projects collaborate, adapt ideas from each other, and strive for portability and interoperability so that compatibility problems do not exist and so that users have the greatest freedom in terms of both hardware and software; but I guess that may be too much to ask

Re: Chimera Linux

#178
post #146

If someone is also looking for more details on dinit, here is the repo: https://github.com/davmac314/dinit

Thank you. Could you give summarize the main differences in approach between dinit and S6? I find S6 a stellar init system, so I'm interested.

The author has a high-level overview doc here: https://github.com/davmac314/dinit/blob/master/doc/COMPARISO...

Now, of course, with service management the devil is in the details, so before you use something as system-wide init, I find it useful to ask yourself: How do they restart services? Do they give up at some point? How do they notify administrators of failures? Do they detect crashloops? How configurable is the logger? What CLIs are there to debug the state of the system (which service was first to fail, where is its definition)? How to make ephemeral or parametrised services? How to add pre-start, post-start, pre-stop, post-stop hooks? Can you use environment variables in your commands, and where do they come from?

I don't think you can find answers to some of these questions in docs. Once you do learn the answers, they may be disappointing - I indeed found myself quite disappointed in systemd after having to debug many failed-to-boot machines. With s6, I never run it, but there are few choices that raise eyebrows, e.g. the restart delay is hardcoded in source code to be 1 second. With dinit, I have yet to finish reading all its manpages, but at least timeout and restart policies are configurable.

Re: Chimera Linux

#179
post #162

Even though I do not fully agree, I think this project has one of the most mature, nuanced, and balanced takes on systemd I have seen yet [0]. [0]: https://chimera-linux.org/docs/faq#what-is-the-projects-take...

A Bit OT, but is there something like the contrary to this - a distro actually fully embracing systemd components? I have spent lots of time with Linux servers. I’m also very pragmatic. systemd has made lots of stuff easier, is highly configurable, maintained and regularly updated, and properly standardised. I acknowledge people who have a need to feel in control or something, but I just want reliable computers that…

Not really. Which is frankly odd because Poettering definitely has strong opinions about how a distro should be distributed and structured... Curious why he doesn't just do it himself.

Re: Chimera Linux

#180

Earlier quoted context omitted.

Yeah pretty interesting. Personally I never liked systemd because I just didn't see the need for it and I don't like changes forced on me. I'm not a team player and I hate being told to 'just go with the flow'. The whole idea of FOSS for me is to be in control of my computer. Alnd I just don't like the character of Poettering and yes that matters. For a much more extreme example look at ReiserFS. It was totally shitc…

> They don't have the push to constantly changing stuff around for the sake of things like systemd and net-tools. I assume you forgot a word here and intended to write "changing stuff around for the sake of changing things"? I just want to set the record straight: systemd solves the startup-dependency problem. Yes the old default init scripts worked fine for your system (that probably had very few services installed…

I don't really agree, because it works totally fine on BSD. I do think the Linux init system was a bit less advanced than BSD's though.

PS: I don't like Wayland either for that matter :) At all. My ideal would be an X12 with the X-Windows foundations but modified for modern realities.

Post reply on HN