Fedora 18 console screen res, help me understand systemd

Lonnie Olson lists at kittypee.com
Thu Mar 28 16:49:19 MDT 2013

On Thu, Mar 28, 2013 at 4:19 PM, Daniel Fussell <dfussell at byu.edu> wrote:
> And so they chased after an OS that was once again dumbing down the
> interface to meet the intelligence of it's "users".  This might work for
> the general masses.  But someone that seeks enlightenment and truth
> wants an interface the pulls them up, not dumbs them down.  We seek an
> intelligent interface that improves out lives.  Dumbing down the
> interface just dumbs down the user.  And those users are too dumb to
> switch to a better OS anyway, so why chase it.  Give people with the
> desire to have a free mind find their way out of the matrix a place to
> go.  Do shove them back in an call it an improvement.

I strongly disagree with this sentiment.  These new desktop interfaces
are not dumbing down the interface.  They add tons of features that
previously required tons of third party addons and configuring to
work.  Features that increase productivity.  I for one couldn't live
without built-in instant search for apps, files, etc.

They do remove some of the insane level of options in the "control
panels" that more often than not tends to break stuff
(http://limi.net/checkboxes-that-kill).  However they continue to
provide a rich set of customization through hidden settings or
extensions for advanced users.  I would argue if you truly only cared
about customization, you would be using KDE, E17 or some other strange
window manager.

Yes, people who get stuck in their ways may experience some friction
in the change, but those people will have the same issues with *any*
change.  (Reads ahead about systemd and sees it's confirmed :)  )

> Systemd rings of the windows registry to me (different product, but same
> user reaction).  I suspect it will be a total flop.  Yes it's nice to be
> able to start multiple processes at once.  Wait, the current sysvinit
> can do that now.  "Oh, but it doesn't make things faster on boot."
> Right, because booting is mostly a drive intensive operation, regardless
> of init system; parallelism will only scatter the heads more, further
> delaying startup.

systemd is not just about booting faster, it's about adding tons of
init features that are sorely needed, and those that have been in use
through ugly hacks before.  Dependencies, socket activated daemons,
process watching, etc.  Also, creating new services/jobs is much
easier than trying to hack together a custom init script.

Yes, there will need to be some initial learning to be done to
understand the new ways of doing things, but it's not much more than
the initial learning you had to do with SysV Init scripts + distro
specific handling of them (chkconfig vs update-rc.d)

More information about the PLUG mailing list