Re: Four people decided the fate of debian with systemd. Bad faith likely
On Sunday, March 02, 2014 01:28:57 PM Doug wrote:
> On 03/02/2014 02:02 AM, Scott Ferguson wrote:
> > On 02/03/14 16:53, yaro@marupa.net wrote:
> >> On Sunday, March 02, 2014 04:25:13 PM Scott Ferguson wrote:
> >>> On 02/03/14 11:28, Ralf Mardorf wrote:
> >>>> On Sun, 2014-03-02 at 10:55 +1100, Scott Ferguson wrote:
> >>>>> Here's mine:-
> >>>>> troll elsewhere - try rabbleRus.org or LetMeTellUWhat2Do.mob
> >>>>>
> >>>> :D
> >>>>
> >>>> We Arch users made a poll. Even if more users would have been against
> >>>> systemd, the developers would have switched to systemd, but most users
> >>>> wanted systemd. We, around 49% and me were against systemd, but around
> >>>> 51 % were pro systemd. Nowadays it makes live easier for all of us who
> >>>> use several different distros, when _all_ or at least the most
> >>>> important
> >>>> distros will switch to systemd. To discuss pros and cons of systemd a
> >>>> time machine is needed, to go back more than 3 years ago. To discuss it
> >>>> in 2014 is a little bit to late.
> >>>
> >>> Same with Debian based on what I read, the vote was fairly evenly split,
> >>> which is why it went to the Technical Committee, who were also fairly
> >>> evenly split.
> >>
> >> Which probably demonstrates why there's no hidden agenda going on
> >> surrounding systemd and there were legitimate reasons why it was finally
> >> chosen.>>
> >>> My concern is that it's a divisive issue that would be tempting for
> >>> third parties to exacerbate and exploit. Commercial software vendors,
> >>> and the companies that do their "marketing" and "public relation" might
> >>> want to take advantage of the situation to reduce the market share they
> >>> lose to Debian (and Linux as a whole). It wouldn't be that far from the
> >>> sort of dirty tactics they've employed in the past.
> >>
> >> Definitely reasonable concerns, though to be honest, Linux's detractors
> >> would have looked for something else to latch onto if systemd wasn't
> >> divisive enough.
> >
> > As well as?
> >
> >> In a few more years I imagine most people opposed to systemd won't
> >> have a problem with it being there after all after using it for a bit.
> >
> > I'd be very surprised if it wasn't modified to suit the needs of the
> > majority of developers - and they tend have the same itches as the
> > "users", just slightly less conservative about their "needs". But I'm
> > not a futurist.
> > Though I did try voting conservative for a change - not surprisingly I
> > was disappointed ;p
>
> Has it occurred to anyone that the devs keep making changes to things
> that work perfectly well, because if they didn't, they wouldn't have
> anything
> to do, and might be out of a job?
>
> --doug
Well, while I wouldn't rule out change for change's sake, I do personally
believe this was an actual needed change, between how inefficient and
problematic initscripts can be to how badly Linux needs an actual system
manager capable of unifying configuration, device management, and service
control...
Not to mention sysvinit has even been stated by it own upstream maintainer
that it's become a trouble to upkeep.
Sure, systemd has its flaws (While I like the journal, there are downsides to a
binary-based log when your system is screwed up and your only resource is a
LiveCD. I don't know if there's a way to read the journal outside the system
that created it.), but ultimately between our choices: Stick with SysV,
Upstart (Which takes an everything and the kitchen sink approach to its
dependency startups and encourages complexity.), and OpenRC (Which utterly
misses the reasons why SysV needs replacing.), I'd choose systemd.
The only arguments I've seen against systemd, at least in this thread is
either "it's change, and change is evil" and "Red Hat/Lennart did it, so it
must be bad." I think a lot of the resistance seems grounded in an irrational
hatred of corporate involvement in Linux. IT's VERY irrational given that a
huge portion, if not most of, the kernel itself is corporate code from
companies like Red Hat, IBM, Intel, Motorola, Google, HP, and even
Microsoft...
A significant portion of the drivers in the kernel tree are, themselves,
provided by the company that made the hardware in the first place. Drivers for
Intel GPUs on Linux ARE the official Intel-provided driver and are part of the
tree.
Strip away all corporate contributions and support and Linux really IS a hobby
OS no one can use for anything.
Conrad
Reply to: