On Thu, 11 Jan 2018 14:35:58 -0500
Hendrik Boom <[email protected]> wrote:
> On Wed, Apr 26, 2017 at 07:57:54AM +0100, KatolaZ wrote:
> >
> > Technical motivations have always been too feeble to cause
> > revolutions. We are not here just because systemd is technically
> > flawed (as we think it is). We are here because we feel that the way
> > in which sytemd has been forced down out throats is a violence
> > against the entire community. And the community side cannot be put
> > in a whitepaper ;)
>
> It's the way that it has been forced down out throats that lets it be
> viable even though technically flawed.
Which comes around to the original question about free software: Is
SystemD really free software?
Obviously, by the FSF definition of free software and the current
definition of open source, yes it is. By the letter of the law, yes it
is. Spirit of the law? I'm not so sure.
But these definitions were made in the 80's (free software) and 90's
(open source), long before the application of huge amounts of money to
write free software, and longer before anyone understood the results of
huge money contributions to free software.
I don't think the definitions of free software and open source
anticipated a form of obfuscation so powerful that a simple computer
program couldn't de-obfuscate it. Probably because they didn't
anticipate success to the point where a commercial company would devote
five or ten top notch programmer salaries to creating and maintaining
the obfuscation method.[1]
My point in the preceding paragraph is easy to argue against. Is X any
less complex and obfuscated than SystemD? Less complex, probably not.
But I get the feeling that the complexity of X is the result of it
reflecting the MUCH more complex problem domain than that of an init
system (even one with a "toolkit"), and also the fact that X is decades
old and seriously in need of a total refactoring (but a refactoring
without any systemd dependency). SystemD's obfuscation is deliberate.
One could also bring up the fact that a determined group of programmers
could tame/refactor systemd. Well yeah, but would they go through all
that work just to end up with something much less useful than the s6
init system?
GPL3 recently broadened the definition of (its take on) free software
by declaring software patents to be nonfree unless given to the
community (do I have that right?). Perhaps some day, a license will
address the issue of deliberate obfuscation, although this is much more
iffy and the line is fuzzy.
Just speaking for myself, because of its deliberate obfuscations making
integration extremely difficult, I don't consider SystemD to be free
software.
[1]: This point is credible only if one believes the primary motivation
for maintaining systemd is the obfuscation of Linux to make money
with Linux training and consultation. I have that belief.
SteveT
Steve Litt
January 2018 featured book: Troubleshooting: Why Bother?
http://www.troubleshooters.com/twb
_______________________________________________
Dng mailing list
[email protected]
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
