[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Why focus on systemd?

2014/11/23 2:57 "Lisi Reisz" <lisi.reisz@gmail.com>:
> On Saturday 22 November 2014 16:03:54 Buntunub wrote:
> > I
> > certainly have no qualms about including Systemd in the next stable
> > release, and I actually encourage it, so that people will have time to play
> > with it and come to know and possibly even fall in love with it. I
> > seriously do not understand why this needs to be rushed.
> Systemd is available in the current stable, (see below) and people have had
> time to play and fall in love with it (or not).  So in how many Stable
> versions of Linux are you saying that it should be available, before those
> who wish to do so will have had long enough to play with it?

Well, that is precisely the question.

How many distros are there?

How many different ways have people tried to figure out to do things, leaving their results in insufficiently commented code, here and there, all over the source code tree, in different repositories all over the world?

How many different algorithms is systemd going to have to fit itself nicely to, before it quits breaking _existing_ infrastructure?

What do you think Miles should have done to avoid that all-nighter? (Yes, his example was udev, not systemd, but the problem of hidden things breaking is the same.)

Well, the truth about that last question is (Sorry, Miles), you can't avoid those. But overly optimistic thinking on his part doesn't excuse overly optimistic thinking on anyone else's part.

And that's the reason I keep saying systemd deserved a parallel, internal fork:

"It works for me!" does not mean "It works for you!"

(Not picking on you, Lisi, you just asked the right question at the right time.)

Joel Rees

Reply to: