[HEADS-UP] systemd for F14 - the next steps

[Matthew Miller]

On Thu, Jul 22, 2010 at 03:52:06PM +0200, drago01 wrote:
> >> > That's what we call a successful transition. Now, we can incrementally
> >> > introduce improvements over the next few releases.
> >> Once you start doing that people will cry because it is different from
> >> what they are used too (does not matter if the change is for the
> >> better or worse).
> > So, let me ask a question: why do you think people have that response?
> Read this thread? Or any other one that introduces any major change?

I'm sorry; let me rephrase. What do you think it is that *causes* that
response in people?

> No, but even if would be very unlikely ... producing books costs time
> and money you don't do it because someone said "sometime we will start
> using foo".
> If anything books document existing stuff (and most of the time lag
> behind), not what is planned in X years.

I've been the technical editor on several Linux books, and done technical
review for others. While they do document what exists, having changes sells
new editions, so keeping current is a priority.

> > Leadership means making careful, well-conceived decisions. Otherwise,
> > it's not leading, it's charging around blindly shouting "follow me!".
> Yeah, systemd seems to replace a part of the system which has been
> pretty much stagnant for ever and implements a  lot of interesting
> concepts.
> On the other side we have "it isn't 100% like sysvinit" ...

That isn't, actually, a fair representation of the "other side".


> > Instead, you seem to be trying to argue that change comes for free.
> No I am just saying that a change isn't bad because it is a change.

"Bad" is a value judgement that I think is unhelpful here. It's not that
change is _bad_, it's that change isn't cost-neutral. When you want to
change something, you're starting at the bottom of the hill, not at a
neutral middle ground.


> > (And
> > that those who have to pay this cost are "crying".)
> Everyone has to pay this cost and everyone gets something in return.

And the way you present this as an _overall win_ is by emphasizing the
returns and decreasing the costs.

What I hear is discounting the costs as real, rather than actually trying to
decrease them. This makes enemies out of your would-be supporters.





-- 
Matthew Miller <mattdm at mattdm.org>
Senior Systems Architect -- Instructional & Research Computing Services
Harvard School of Engineering & Applied Sciences