Originally posted by Daktyl198
View Post
Announcement
Collapse
No announcement yet.
GNOME 3.11.2 Has Many Changes
Collapse
X
-
Originally posted by Daktyl198 View Post"no one offers an alternative" I'm currently using Mint 16 which uses Upstart + Cinnamon. A perfectly usable combination, thank you very much. If udev starts actually using Systemd functions (instead of just being packaged to depend on it), we have eudev.
If that combo ever goes away (which I doubt) am perfectly fine and happy using Systemd, I'm not a hater of the software itself (it does provide quite a few things I like and writing a script for it was super easy).
Here's a couple real questions I have though: What was the reason for Systemd's creation? I mean, I know it was a "pet project" and all, but is there any reason besides that? Then, why did it become so big so fast when it was created to solve the same problems as upstart and other stuff. Why didn't everybody (intel, etc) jump on the upstart train? Because Red Hat backs Systemd?
Upstart doesn't come close to offering what systemd does. If you want to know why it was created here's the announcement link: http://0pointer.de/blog/projects/systemd.html
Things to look for in that post, imho, are: keeping track of processes, keeping pid 1 small, and on upstart.
Upstart simply doesn't cover quite the same area, or do it in as well, as systemd. One of the big telltales of this is that, apparently, ubuntu has still failed to convert a very sizable chunk of their shell scripts to upstart files. Again, apparently, upstart has issues with both properly tracking processes, and doesn't offer anywhere close to the backwards compatibility with sysvinit that systemd does (which, itself, isn't perfect, but does an awfully good job just working with old scripts).
There is a reason so many people from so many different groups have converged around systemd. It actually solves problems, and is architected extremely well.
BTW, RH actually use upstart in rhel 6, so they backed the project for awhile.
Comment
-
Originally posted by pingufunkybeat View PostNah.
The strength of the Linux ecosystem is in its diversity. That's also what I like about it. I'm not interested in a GPL-ed Macintosh clone.
Linux would not be interesting if you couldn't tinker and try different options.
Certain things should JUST WORK. I want acpi to be part of that. If you want to fiddle with your kernel config, init daemon, power management framework, language runtimes, be my guest, b/c I'm positive you will continue to be able to do that for a long time, but you'll probably be compiling them, and carrying lots of patches, but you can certainly do it.
The HPC folks do it all the time.
Comment
-
Originally posted by Daktyl198 View Post"no one offers an alternative" I'm currently using Mint 16 which uses Upstart + Cinnamon. A perfectly usable combination, thank you very much. If udev starts actually using Systemd functions (instead of just being packaged to depend on it), we have eudev.
Comment
-
Originally posted by Daktyl198 View PostWow. Condescending much? Not that it would matter to you, but I'm learning C to eventually contribute to Mesa and Cinnamon (My favorite DE at the moment). It's kinda hard to "help out" when you don't know enough of the language.
P.S. Always nice to see how a member of the Gnome release team treats Gnome users who can't contribute code but have criticisms :/
Comment
-
Originally posted by liam View PostMaybe Linux shouldn't be interesting, in that sense?
Certain things should JUST WORK. I want acpi to be part of that. If you want to fiddle with your kernel config, init daemon, power management framework, language runtimes, be my guest, b/c I'm positive you will continue to be able to do that for a long time, but you'll probably be compiling them, and carrying lots of patches, but you can certainly do it.
The HPC folks do it all the time.
I use Linux a lot. It is really nice that you can adjust things. But adjusting will always be possible: you have the source, you tinker with every part, you can contribute whereever you want, you can build things exactly like upstream does. Tinkering forever: cool, but it also has to work. Then if it works, it can be fun to take apart and figure things out. But saying that the main purpose is tinkering? That's not why a lot of people are contributing. Seems related to: http://www.islinuxaboutchoice.com.
Comment
-
Originally posted by bkor View PostSeems related to: http://www.islinuxaboutchoice.com.
Really?
This one person declares that Linux is not about choice and you use it as an absolute fact?
Wow, is that the style all discussion works in Gnome?
Comment
-
Originally posted by Vim_User View PostI see. Some Gnome related person sets up a website with a big "NO" on it and somehow it is true for all and everything, just because he says so.
Really?
This one person declares that Linux is not about choice and you use it as an absolute fact?
Wow, is that the style all discussion works in Gnome?
Comment
-
Originally posted by RahulSundaram View PostWrong. Adam Jackson is a Xorg developer and doesn't work on GNOME.
Originally posted by bkor View PostAgree fully.
Also I've seen a NO site that looks better than that: http://shouldiusetablesforlayout.com/
Originally posted by finalzone View PostThat sounds like a misinformed post. Systemd currently depends on udev.
Originally posted by liam View PostUpstart doesn't come close to offering what systemd does. If you want to know why it was created here's the announcement link: http://0pointer.de/blog/projects/systemd.html
Things to look for in that post, imho, are: keeping track of processes, keeping pid 1 small, and on upstart.
Upstart simply doesn't cover quite the same area, or do it in as well, as systemd. One of the big telltales of this is that, apparently, ubuntu has still failed to convert a very sizable chunk of their shell scripts to upstart files. Again, apparently, upstart has issues with both properly tracking processes, and doesn't offer anywhere close to the backwards compatibility with sysvinit that systemd does (which, itself, isn't perfect, but does an awfully good job just working with old scripts).
There is a reason so many people from so many different groups have converged around systemd. It actually solves problems, and is architected extremely well.
BTW, RH actually use upstart in rhel 6, so they backed the project for awhile.
Comment
Comment