init observations

[Jeff Spaleta]

On 11/15/05, Lamont R. Peterson <lamont at gurulabs.com> wrote:
> I think that what he meant by laptop userbase vs. desktop userbase is valid
> because *most* "desktop" users don't care about suspend.  Even if they see it
> available, they probably won't use it.

User education problem.... which only serves to re-enforce the status-quo.

> I think you are right, though, Jef; desktop systems, set-top boxes, and so on
> and so on, would definitely benefit, too.  But I have to wonder, since the
> developers who are going to work on both possible solutions are probably not
> the same people, why not just do both?

The context in this discussion is.... does the parallelization that
initng provides actually improve boottime or does it not. My argument
is.. compared to the other technical "wins" with changing init systems
the boottime "benefit" is speculative and should not be the focus of
the decision at hand. Hell, initng might actually cause a boottime
increase for the default case.

I'm all for continued optmization of the boot process, the creation of
the bootchart tool and the optmization built on the use of that tool
has been a good thing{tm} and I'm sure more progress will be made on
that front regardless of which init system is in use. But, I think its
ill-advised to make the focus of the discussion of changing the init
system  centered around boottime enhancements which may or may not
materialize. As a matter of priorities  associated with the specific
issue of changing the init system... boottime ehancement should be the
last concern.

-jef"most desktop users dont care about linux.. even if they see it
available they won't use it.... but here we are anyways.. attempting
to educate and to make better solutions "spaleta