3 ms·

The fundamental problem here is that desktop Linux users, who have absolutely had a bad experience over the decades, have gotten so tired of incompetently execu

The fundamental problem here is that desktop Linux users, who have absolutely had a bad experience over the decades, have gotten so tired of incompetently executed half-assed solutions that they now grab at any other solution with the fervor of a drowning victim. Why have they gotten to this point? Because the people working on desktop have generally been the D-team. The A-team is working on kernels and infrastructure, the B-team turns out packaging, and the D-team has been coming up with hair-raising stuff like gnome, 'desktop bus', pulseaudio, enlightenment. So this weird little monoculture has emerged in which no other concern matters except that they get rescued from drowning -- that their boot times (!?!?) go to 5 seconds from 10, that hot-plugging USB drives (!?!?!?!) be wrapped in with the init system if it helps at all, that DHCP (!!??!?!?!?!!) and core files (!!!!!!!!) and all logging (!!!!?!?!?!?!?!?!) be taken over in the name of helping even 1% on the core experience. But at the same time, two basic axioms of reality still bear: 1. Linux on the desktop is irrelevant and will never win, both in terms of the Linux story and in absolute terms; and every scrap of effort that goes into it has been objectively wasted for two decades and will continue to be wasted. 2. Linux everywhere else has been working fine for longer than most of the desktop people have been alive. following from (2), the people who run servers recognize that although systemd has some good ideas, it forcibly bundles together a grip of super bad ones too, and is clearly implemented by people who don't understand systems outside of the desktop. The attempt to turn Debian back into a distribution for everyone, rather than a small collection of desktop people, is completely valid and sensible. That said, it's totally clear why the desktop people want it this way, and why they have such a misguided emotional investment in their terrible new windows emulator.

pyre 12y ago
> pulseaudio To be fair, some of the ideas behind pulseaudio are good: - Unify all of the different audio systems that existed: OSS, ALSA, esound daemon, etc... - Network accessible sound. Years ago I was able to setup pulseaudio to stream audio from my HTPC to my laptop where I had headphones plugged in. The process to do this wasn't entirely user-friendly, but prior to pulseaudio, people would setup yet another audio subsytem that every app would then need to support on a case-by-case basis.
EmanueleAina 12y ago
As the existence of CoreOS proves, much of systemd benefits are not desktop-oriented, and I can assure you that many systemd developers/maintainers are not desktop developers but experienced sysadmins/developers. Thas said, while I may agree with the spirit of the GR, like I may agree with the existence of the Hurd and kFreeBSD ports, I wonder who will do the work needed to patch and support other init systems, given that up to now no-one bothered putting some actual work instead of pretending others should do it. At least the Hurd and kFreeBSD porters were kind enough of carrying out most of the effort themselves, even if unfortunately I don't think it was sufficient. (And no, please, systemd has nothing to do with Windows, please avoid such baseless comments which only manage to start useless flames)

pyre 12y ago

> pulseaudio To be fair, some of the ideas behind pulseaudio are good: - Unify all of the different audio systems that existed: OSS, ALSA, esound daemon, etc... - Network accessible sound. Years ago I was able to setup pulseaudio to stream audio from my HTPC to my laptop where I had headphones plugged in. The process to do this wasn't entirely user-friendly, but prior to pulseaudio, people would setup yet another audio subsytem that every app would then need to support on a case-by-case basis.

EmanueleAina 12y ago

As the existence of CoreOS proves, much of systemd benefits are not desktop-oriented, and I can assure you that many systemd developers/maintainers are not desktop developers but experienced sysadmins/developers. Thas said, while I may agree with the spirit of the GR, like I may agree with the existence of the Hurd and kFreeBSD ports, I wonder who will do the work needed to patch and support other init systems, given that up to now no-one bothered putting some actual work instead of pretending others should do it. At least the Hurd and kFreeBSD porters were kind enough of carrying out most of the effort themselves, even if unfortunately I don't think it was sufficient. (And no, please, systemd has nothing to do with Windows, please avoid such baseless comments which only manage to start useless flames)