Systemd's Networkd Now Supports Bridging

[ceage]

BSDs have their issues. But having a corporation that tries to hijack the whole system by writing an init replacement that is incidentally a dependency for everything else is not among those issues.

How exactly are they doing the hijacking? By writing more copylefted free software for use in a copylefted free software environment?
There is no hijacking. That's exactly the kind of behavior that the GPL prevents.

[prodigy_]

There is no hijacking. That's exactly the kind of behavior that the GPL prevents.

/sigh No, unfortunately, as of now it doesn't. Logind, logger and networkd are just the beginning. Continue this discussion in 5 years, shall we?

[zanny]

BSDs have their issues. But having a corporation that tries to hijack the whole system by writing an init replacement that is incidentally a dependency for everything else is not among those issues.

This is like saying that Red Hat, Intel, IBM, and Novell are all hijacking kernels by sponsoring a third of kernel development.

Surprise, a lot of the best critical infrastructure software is written when you have corporate backers who need stability. It takes a lot of testing and infrastructure to make a stable init system, or stable kernel, or stable IPC, etc. So you find that the best ones emerge from those where the dev team is paid for their work, predominantly. Of course Linux / systemd / etc all get tons of contributions from hobbyists, but they are hosted and maintained principally on a paycheck.

[peppepz]

Surprise, a lot of the best critical infrastructure software is written when you have corporate backers who need stability. It takes a lot of testing and infrastructure to make a stable init system, or stable kernel, or stable IPC, etc. So you find that the best ones emerge from those where the dev team is paid for their work, predominantly. Of course Linux / systemd / etc all get tons of contributions from hobbyists, but they are hosted and maintained principally on a paycheck.

I think that this statement is too general. Compare Android (fully developed by a single corporate backer) with the Linux kernel (a great deal deal of paid developement, but also with hobbyists and academic contributions, and managed with a more bazaar-like approach). The former has lower quality, introduces APIs very frequently, and breaks compatibility at the same pace.

[Rallos Zek]

BSDs have their issues. But having a corporation that tries to hijack the whole system by writing an init replacement that is incidentally a dependency for everything else is not among those issues.

Couldn't said it better even with a translator.

[fscan]

BSDs have their issues. But having a corporation that tries to hijack the whole system by writing an init replacement that is incidentally a dependency for everything else is not among those issues.

So, providing new features that a lot of projects want to use is a bad thing now? Hell, they even provide stable dbus interfaces for most things to be easily replaceable.

btw, i love your logic: A introduces new feature, B, C and D want to use it because it's the only viable option out there -> A is the bad guy

[mrugiero]

I read that they target basic stuff and not the advanced things you can do with NM.

What would be considered basic and what would be considered advanced? If it works as a replacement to the interfaces file, it is enough for me.

BSDs have their issues. But having a corporation that tries to hijack the whole system by writing an init replacement that is incidentally a dependency for everything else is not among those issues.

How this hijacks the system? For a start, is it a dependency for everything else? If so, who decided it would be a dependency for everything else? Did systemd developers decided that GNOME (as that's the only ones I can think of really depending on systemd) would be depending on them, or did GNOME folks decided this?

/sigh No, unfortunately, as of now it doesn't. Logind, logger and networkd are just the beginning. Continue this discussion in 5 years, shall we?

What should happen in 5 years? Systemd is supposed to become closed source in 5 years, or something like that? Is there any copyright assignment that allows them to do so? Because as long as there are outsiders contributing, they'll need their copyright to even think of relicensing.

I think that this statement is too general. Compare Android (fully developed by a single corporate backer) with the Linux kernel (a great deal deal of paid developement, but also with hobbyists and academic contributions, and managed with a more bazaar-like approach). The former has lower quality, introduces APIs very frequently, and breaks compatibility at the same pace.

Yes, and that's as simple as "it follows a cathedral approach, so it doesn't need to be good, it just needs to do what its backer wants". When you have a group of people/companies, they will push each other to keep it flexible enough to keep being useful for them. Android is consumers-only, so they don't need huge stability, and is hegemonic enough to be able to break compatibility and APIs frequently, as most developers are forced to go with the flow or switch jobs.
If Qt were to break APIs every few months, for example, most developers would just give up on their portability desires, and just target Windows or Mac OSX, for example. While Android breaking it means, well, you have to adapt to those changes every few months, as Android and iOS are the only really big players on mobile, and you need to sell your app to somebody. So, Android has a captive market and doesn't need to comply with other companies requests of stability, naturally they will get lower quality, mostly because they don't need it to be any higher (and higher quality is also more expensive).

[prodigy_]

This is like saying that Red Hat, Intel, IBM, and Novell are all hijacking kernels by sponsoring a third of kernel development.

Begone, troll. Kernel development is not controlled by corporations and hijacking isn't about how much code you contribute.

[prodigy_]

What should happen in 5 years? Systemd is supposed to become closed source in 5 years, or something like that? Is there any copyright assignment that allows them to do so? Because as long as there are outsiders contributing, they'll need their copyright to even think of relicensing.

And how's that an obstacle? They don't need a copyright on systemd, they don't need to sell it. Their goal is to control the ecosystem. Want an example? MS never sold browsers (not separately from Windows anyway) yet the popularity of IE basically forced web-developers to ignore standards for years.

And systemd is LGPL, not GPL. LGPL is fine for non-critical code but init is critical. If systemd wins, what can stop RedHat from adding proprietary "extensions" and suddenly making all of us who don't pay them second class citizens?

[finalzone]

Corporation? You do understand that systemd-networkd is developed by Tom Gundersen, an Arch Linux developer, and the dhcp stuff is written by Intel? Also systemd as a whole is collaboratively developed project with ever increasing number of contributors.

To back up this point:

http://www.ohloh.net/p/compare?project_0=Upstart&project_1=OpenRC&project_2=systemd#