:: Re: [Dng] Combatting revisionist hi…
Forside
Slet denne besked
Besvar denne besked
Skribent: Neo Futur
Dato:  
Til: t.j.duchene
CC: dng@lists.dyne.org
Emne: Re: [Dng] Combatting revisionist history
>> This excellent analysis of the systemd debacle was just posted over on FDN. Should be required reading IMO. Enjoy!
>>
>> http://forums.debian.net/viewtopic.php?f=20&t=120652&p=570371
> If something replaces init, it is by definition "an init system".
> Whether it does more or less than the previous init is immaterial to
> that simple fact.


It does so much more than just the init system . . . everyday more
userspace features are gobbled by systemd . . .
The more time passes, the more I agree with this post, the init
process is now just one feature, the one ring needed to eat them all .
. .

with time this looks more and more like a systemd-linux OS, than a gnu/linux OS

> What I find interesting about the whole affair is that for all the
> arguments for and against systemd, no one seems to really bring out the
> one argument that really makes a difference. All of Unix follows POSIX
> in some major measure. The init system is not covered by the POSIX
> standard.


concerning posix, afaik theres a problem at the libc level , libc is
posix and glibc is much more than posix, systemd pushes even more
glibc as the new standard against posix, afaik you wont be able to run
systemd on a non glibc based system, like the uclibc gentoo stages,
thats one more endangered alternative ( not using glibc ).

with time this looks more and more like a redhat-glibc-systemd-linux
than a gnu/linux OS


> There have been other replacements for init in the past. The fact that
> Linux is going through this particular "growing pain" that other Unixes
> already have is nothing new and certainly not the end of the world.


how many replacement to the init system have slowly eaten allthe
userspace features, like logind, syslog, pam and so many others ?

> Devuan is proof of that diversity can exist and should. So is Gentoo
> and some others out there.


sure

> If I had to point the finger at Linux's greatest failing, it is the
> expectation that users want others to do all of the work compiling and
> packaging and then they want to complain that that someone else made a
> choice that they didn't like.

well i wont say that as a gentoo user ;)