Gentoo Gets GNOME 3.30 Running Without Systemd

[Veerappan]

As a fellow Digital Personal Workstation owner, I salute you. o7

Mine currently doesn't have an OS, and I was just going to put OpenVMS on it, but it'd be cool to try Linux as well. It would be cool if you wrote up a guide when you're done.

This would be where you might want to get started: https://wiki.gentoo.org/wiki/Handboo...l/Installation

[Veerappan]

Which required kernel features are missing on Alpha?

I know this is an ancient thread at this point, but I have to at least admit that I was wrong here. I managed to get systemd working on Gentoo/Alpha (along with Gnome Shell using a radeon HD 5400-series PCI card).

[aht0]

I think unapproved posts "going away" for premium members, at least for posts treated as potential spam, would be nice.
-----

Wow. It's arrogant to spend your own time on stuff that interests/pays you, rather than wearing yourself out selflessly making your project support every Tom, Dick, and Harry who comes to call?

No. It's arrogant to think that everyone should think like you/me/whoever else does, like uniform sheep. It is natural that different people like different things. It's arrogant to subtly enforce it by writing a software, which by abusing the popularity of the particular devs "love child" actually does it. It goes against basic courtesy of "live and let live", it becomes "live like I think you should" - enforcing one's will, be it subtly or intentionally. In other words, it becomes sort of software dictatorship. Like Windows actually - You are more or less forced to use it when you buy new OEM machine and you have to take special effort to get rid of it.

You are wholly ignoring the fact that said dev or devs are not the sole human beings in this flippin' Universe but their actions actually affect others to a degree. In real life, as a society, we generally and for the most part try to act following certain rules of courtesy. In virtual world, it should be same but for some reason when face-to-face interaction falls off, people let their true character show. So, yes, arrogance." I know the best". Which time to time shows through even systemd's bug tracker.

I'd say it's a perfectly good analogy because:

1. As an end-user, FreeBSD and Windows 98SE are basically the same: Operating systems which are niche at this moment which I use. (I run a FreeBSD-based router and I run a Windows 98SE retro-gaming PC which, like pre-online consoles like my N64, is airgapped while, like other PCs, is something I can and do develop for.)
2. From the perspective of the systemd developers, the BSDs are like Windows 9x. Some people use them, and it's technically possible to support them, but their market share is tiny in relevant sectors, it's far too much work for too little effort, and there's no "it's my hobby" factor to motivate them despite that.

Operating systems which YOU use. Analogy is perfect for YOU. Again, now I have to point finger at you - arrogance. There are billions who use only windows. For them, Linux is niche.

I actually followed train-of-thought of "End of Life". BSD's are developed actively, ergo your analogy is wrong in that particular aspect.

As long as it remains open-source and configurable in the ways I desire, I don't begrudge them this. In fact, I'd say this is somewhat overdue. Linux has had a history of being crippled by a community with a dogmatic fetishization of variety for variety's sake far in excess of what is healthy if you want to develop a platform.

(PulseAudio may be a pain in the ass, but the chart of what the audio stack could wind up looking like before it is infamous.)

[/QUOTE]
You don't begrudge them just until it actually stops you from doing or using something you need or want. Then you start hating it, because working against it's intentionally created barricades is so flippin' annoying for all "dissidents".

[ssokolow]

Pinging Michael. Another linkless un-approved post.

[ssokolow]

(1)Like I said, sheer arrogance.

Wow. It's arrogant to spend your own time on stuff that interests/pays you, rather than wearing yourself out selflessly making your project support every Tom, Dick, and Harry who comes to call?

(2)And I was not talking about supporting long-dead operating systems through compatibility layer of an equally EOL'ed operating system. (Can't call WinXP "dead", US military is still using it, they pay heavy cash to MS for updates). Wrong analogy here.

I'd say it's a perfectly good analogy because:

1. As an end-user, FreeBSD and Windows 98SE are basically the same: Operating systems which are niche at this moment which I use. (I run a FreeBSD-based router and I run a Windows 98SE retro-gaming PC which, like pre-online consoles like my N64, is airgapped while, like other PCs, is something I can and do develop for.)
2. From the perspective of the systemd developers, the BSDs are like Windows 9x. Some people use them, and it's technically possible to support them, but their market share is tiny in relevant sectors, it's far too much work for too little effort, and there's no "it's my hobby" factor to motivate them despite that.

I am predicting now: over time, GNU Linux would become systemd/Linux, tendency to distros become more and more like each-other is already visible, thanks to systemd - main difference in future would be mostly cosmetic - artwork and packages included. Enjoy your Lindows.

As long as it remains open-source and configurable in the ways I desire, I don't begrudge them this. In fact, I'd say this is somewhat overdue. Linux has had a history of being crippled by a community with a dogmatic fetishization of variety for variety's sake far in excess of what is healthy if you want to develop a platform.

(PulseAudio may be a pain in the ass, but the chart of what the audio stack could wind up looking like before it is infamous.)

[cybertraveler]

Then it would be one of two - people either start using more systemdless distros (meaning Linux would effectively split in two)

This is something I think may happen. It looks like it's actually already well under way.

The systemd / Red Hat approach to Linux is quite a big divergence from the prior GNU/Linux approach.

[aht0]

I can see your issues there, but I can also see the viewpoint from the systemd side. Something like "At each level of the stack (systemd, consumers of systemd, etc.), the developers want to make use of the features of the platform they care about and not be held back (1)by what they probably see as 'every little podunk operating system that happens to implement a subset of that'."

(2) I have a nostalgic fondness for Windows 9x, but I don't think it's reasonable to expect the Rust compiler's Windows XP compatibility target to also support it when I put on my hobbyist programmer hat.

(1)Like I said, sheer arrogance.
(2)And I was not talking about supporting long-dead operating systems through compatibility layer of an equally EOL'ed operating system. (Can't call WinXP "dead", US military is still using it, they pay heavy cash to MS for updates). Wrong analogy here.

I am predicting now: over time, GNU Linux would become systemd/Linux, tendency to distros become more and more like each-other is already visible, thanks to systemd - main difference in future would be mostly cosmetic - artwork and packages included. Enjoy your Lindows.

Then it would be one of two - people either start using more systemdless distros (meaning Linux would effectively split in two) OR mass-migrate to more popular BSD's (which have larger developer-base than any current systemd-less distro). There is already quite a lot of "systemd refugees" among FreeBSD user base. Gonna be interesting decade ahead.

[ssokolow]

It's not like UNIX APIs are precisely frozen in time. Most recent POSIX version is POSIX.1-2017. Ignoring POSIX-compliance - which in essence helps solve inter-OS compatibility problems and that among 3rd party apps, is just an typical arrogant "we do whatever the fuck we want"-attitude. Fueled by the sure knowledge that Linux's popularity guarantees they can get away with it.

Remember environment value POSIX_ME_HARDER.

I should probably have phrased that differently. "If UNIX had continued to evolve before adoption inertia made it prohibitive to both make major changes and get them adopted". (eg. If UNIX had continued to become Plan 9 and so on)

You wrong me here, somewhat. I don't feel slightest entitled to free work from others. My main issues with systemd is, it narrowed down "usable for me" Linux distros to bare handful which don't have systemd. WHEN I use Linux at all.
Because, and it might be distributor's problem, whenever I install some distro sporting systemd, it generally would require tinkering to have it working properly.

And it might require it again after random set of updates. As I see it, it's logical result of not following the KISS principle. Overcomplicate something and it will produce errors much more likely

Fair enough. I still think, at the very least, there should be a stronger focus on concrete, addressable problems with systemd that could either be fixed or targeted as distinguishing features by anyone who gets the urge to try to write a viable competitor, and which are too clearly and rationally stated to be easily dismissable.

(eg. my complaint that systemd's design puts too much in PID 1 where it can't be killed and restarted if it wedges, rather than a helper process that PID 1 could respawn if it dies.)

People's behaviour follows the path of least resistance. When there is big corporate entity doing much of the work for them, ofc they'd use it's work, even when they might dislike aspects of it.

Now, lots of anti-systemd folks, me including, are actually users of other UNIX-like OS'es. We dislike systemd for it's virality and negative effect on our OS'es (components replaced by systemd left undeveloped, DE's tightly integrated with systemd components making porting increasingly harder - pretty much systemd is increasingly limiting software both portable and usable on our OS'es).
Developers are finite resource, WHEN they start fixing shit caused by systemd in Linux sphere - who is going to be developing our own OS'es?

Yes, it's easy to say "if you dont like it, develop your own and better solutions". Who single or group of individuals coding during their free time would be able to compete with corporate backed full-time devs? Nobody.

I can see your issues there, but I can also see the viewpoint from the systemd side. Something like "At each level of the stack (systemd, consumers of systemd, etc.), the developers want to make use of the features of the platform they care about and not be held back by what they probably see as 'every little podunk operating system that happens to implement a subset of that'."

I have a nostalgic fondness for Windows 9x, but I don't think it's reasonable to expect the Rust compiler's Windows XP compatibility target to also support it when I put on my hobbyist programmer hat.

[aht0]

McIlroy agrees with you and so do I.

I'm not sure if you've seen the comment in question, but I said before that I think that the problem he describes with manpages stems from UNIX providing insufficient APIs for composing simple components and that, if UNIX development had continued, in addition to Plan9's further extending the "everything is a file" metaphor, we'd have also seen shell pipelines re-standardize on something with basic structure, comparable to what JSON can represent.

It's not like UNIX APIs are precisely frozen in time. Most recent POSIX version is POSIX.1-2017. Ignoring POSIX-compliance - which in essence helps solve inter-OS compatibility problems and that among 3rd party apps, is just an typical arrogant "we do whatever the fuck we want"-attitude. Fueled by the sure knowledge that Linux's popularity guarantees they can get away with it.

Remember environment value POSIX_ME_HARDER.

My issue is with people who are so obsessed with the minutia of a specific interpretation of the UNIX philosophy that they say things like "The UNIX Philosophy is at odds with using something like JSON to pass structured data from program to program" or "The UNIX philosophy forbids the introduction of new APIs and system functionality because it's at odds with the UNIX philosophy's stance on loose coupling and composability".

ehm, agree with you here. UNIX philosophy for me comes down to following the KISS principle. Not staying stuck in 30-50 years old software versions. systemd is anything but simple though.

My two main issues with anti-systemd people on this point are:

1. Do your damn research before complaining about what in systemd is actually tightly coupled and what is just developed under the same umbrella.
2. Rant productively. We're sick of you whining in ways that can only be interpreted as productive if it's assumed that you feel entitled to free work from others.

You wrong me here, somewhat. I don't feel slightest entitled to free work from others. My main issues with systemd is, it narrowed down "usable for me" Linux distros to bare handful which don't have systemd. WHEN I use Linux at all.
Because, and it might be distributor's problem, whenever I install some distro sporting systemd, it generally would require tinkering to have it working properly.

And it might require it again after random set of updates. As I see it, it's logical result of not following the KISS principle. Overcomplicate something and it will produce errors much more likely

As for "it will help me make my work bit easier, so i'll gladly trade more freedom for community against me having to work less", again, you're free to develop an alternative but, apparently, of the options that people took the effort to write, systemd's value proposition is the most appealing to big-name distros and business customers.

People's behaviour follows the path of least resistance. When there is big corporate entity doing much of the work for them, ofc they'd use it's work, even when they might dislike aspects of it.

Now, lots of anti-systemd folks, me including, are actually users of other UNIX-like OS'es. We dislike systemd for it's virality and negative effect on our OS'es (components replaced by systemd left undeveloped, DE's tightly integrated with systemd components making porting increasingly harder - pretty much systemd is increasingly limiting software both portable and usable on our OS'es).
Developers are finite resource, WHEN they start fixing shit caused by systemd in Linux sphere - who is going to be developing our own OS'es?

Yes, it's easy to say "if you dont like it, develop your own and better solutions". Who single or group of individuals coding during their free time would be able to compete with corporate backed full-time devs? Nobody.

[L_A_G]

That's because that is what you do when you make implications on "viral aspects" such as "Normally if you start using one component, like loginD, you're basically forced to adopt the whole thing due to how heavily interconnected all of the parts are" when the reason the various components are interconnected is for a good reason.

The fact that you keep repeating there's a good reason to interconnect practically all of systemD to make it functionally monolithic doesn't actually mean there is one. Sure, when you repeat a lie enough times some people will begin to believe it, but the more observant ones will smell a rat.

Basically your whole complaint boils down to "damn the systemd project for providing something that other developers finds useful". And btw the "he should end himself" we already see by the millions on say Slashdot, it used to exist here as well but thankfully that have more or less disappeared in the last year.

This is exactly what I mean when I say that most of the counter-arguments to criticism of systemD just plain misses the point of the criticism...

Not only do you not address the actual issues people have it with, i.e how it's structured, how it's managed and how the developers handle bug reports and user feedback in general. Nobody ever said that copying the core concepts of launchd from MacOS wasn't a good idea, just like Lennart's previous project where he copied CoreAudio in a similar way was also a good idea on the idea level. No, the issue has always been the execution and the bungling of it is why you have people who would rather stick to init despite being hopelessly archaic similarly to how people stuck to OSS despite it also being pretty damn archaic.

I know you think you got a real zinger on your hands trying to imply that people are telling Lennart to "End himself" when I used it to point out that the criticism that he receives is pretty much completely productive, but I'm not stupid enough to not see right trough it. Lennart is an ass who may have the occasional good idea, but is terrible at project management and even worse at taking feedback and particularly of the negative kind, even when it's totally constructive.