Well, Fedora 14 Will Not Ship On Time

[markg85]

I wonder, when was the last fedora release thst was on scedule? Could that be fedora 8 or 9? (one of the most crappy fedora releases btw)

[Hephasteus]

6 month release cycles are too fast. It turned hp printers from the finest printers on the planet to garbage that is a complete waste of oil making them into plastic crap that you throw in landfills.

[nanonyme]

Long release cycles are too long, short ones are too short and rolling releases unreliable in how well it works. Maybe we shouldn't ever release anything at all? (ps. I'm not responsible for any releases so don't get too agitated of that pun )

[brent]

systemd? Uhm, great. The chaos is complete. "Legacy" init system, upstart and now systemd.

[sirdilznik]

I don't mind Fedora falling behind a week or two behind schedule (as has been the case for the last 2 years plus). To me it's an acceptable delay. It's not like Longhorn, aka Vista, which wound up shipping approximately 3 years behind schedule. Software projects have their original schedules pushed back all the time. Unexpected bugs and complications pop up, it is what it is. Delays of a week or two are quite minor. If we were talking a month or two, then I would say that maybe Fedora needs to rethink it's release schedule or at least the amount of changes per release. But delays of a week or two are just water under the bridge for me. If people absolutely must have the new release earlier they can always get the Beta and get a pretty stable system AND they can help the team iron out the remaining bugs if they wish.

An alternative is going the e17 route of "It'll be finished when it's finished". Well it's been god knows how long and it's not finished yet. (BTW I'm not hating on e17, I like it quite a bit and in fact I'm using it right now)

[V!NCENT]

systemd? Uhm, great. The chaos is complete. "Legacy" init system, upstart and now systemd.

SystemD is very similar to sysv to the admin. SystemD is even compatible with SysV scripts and it uses the same syntax.

SystemD and Upstart will be battling for the title of 'ultimate Sysvinit replacement'

[brent]

SystemD and Upstart will be battling for the title of 'ultimate Sysvinit replacement'

That's exactly what I'm criticizing here.

[susikala]

What I don't get is, even if you need a replacement for SysVinit, why does everyone feel they have to reinvent the wheel and write their own 'better' replacement of it. It's this ego issue that makes FOSS people so unbearable at times.

There's fefe's minit: http://www.fefe.de/minit/

I have been using it on Debian for a while and I have yet to come across problems. Could anyone point me out why it doesn't get adopted by any distro?

[Lynxeye]

Because every new program brings innovations. It's not about having no problem. Systemd brings potentially faster sartup times which could not be achieved with the tradidional init systems.

[V!NCENT]

The reason Systemd is created is because other systems like sysv and upstart are totally different by design. It would be like converting OpenOffice.org into Enlightenment

Lennart has looked at the possibilty of simply improving Upstart but according to him Upstart is flawed by design and even lacks a design vision.

Furthermore SystemD has a totaly different aprouch to lanching and managing daemons. For example dynamically starting and stopping daemons only when needed and keeping track of them by using cgroups. Upstart is just "launch in serialised fashion during startup and prey it all keeps working afterwards." and can't track daemons after doubleforking.