GNOME 3.13.2 Temporarily Depends On Systemd

[Vim_User]

It's not about who writes pieces of code - this is completely irrelevant. It's all about who controls the project as a whole.

No, it is not. You told us that forking systemd would not be viable because developers would have to work intoi 550k lines of code unknown to them. This is obviously wrong. It doesn't matter at all who leads the fork if the developers still know the code.

[Hamish Wilson]

Yeah, reminding people that corporations are not charities equals being a conspiracy theorist. What's next? Does saying that people breath air or that summer comes after spring make one a tinfoil hat wearer too?
/facepalm

Yes, corporations are not charities, and Red Hat is one of the most successful software companies on the planet. They have got there without ever releasing proprietary code, demanding "license fees", or hiding from taxes in the Isle of Mann. They do not need to do any of these things to be successful - Red Hat has such a stable revenue stream that it would be insane for them to do anything to change how they already operate.

As for the fact that they have such a dominating presence in terms of code submissions, project leadership, and Linux technical innovation, that has also been true for the past fifteen to twenty years. I am afraid your nightmare has already arrived, and been in force for almost two decades. That is of course not meant to downplay other contributors, corporate or otherwise, who contribute to the Linux ecosystem - something which you do whenever you claim that systemd is a Red Hat only project.

[prodigy_]

There are a lot of distributions using it and contributing to it, even commercial ones like Suse. Do you think it is written by Red Hat developers alone?

It's not about who writes pieces of code - this is completely irrelevant. It's all about who controls the project as a whole.

[Vim_User]

Forks don't happen by magic. First of all, you need people to maintain a project of such scale. Do you keep up with the news? systemd is 550k lines already. It's horribly bloated but even if the first post-fork goal will be (as it should be) trimming all the fat it's still a lot of work. You can't just comment out everything you don't want. And this is especially true in case of systemd because the developers went out of their way to ensure it's a dependency for as much other stuff as possible.

It's extremely easy to get entangled in the systemd cobweb. But getting out is going to be very complicated and very time consuming.

There are a lot of distributions using it and contributing to it, even commercial ones like Suse. Do you think it is written by Red Hat developers alone?

[TheBlackCat]

Yeah, reminding people that corporations are not charities equals being a conspiracy theorist.

If that was all you claimed then we wouldn't be having this discussion. Or did you forget that this is a forum, and that people can go back and actually see that this has practically no resemblance to your actual claims?

[prodigy_]

You know what the main difference between Red Hat monetization conspiracy theories and Canonical monetization conspiracy theories is? The fact that Red Hat is actually making money (and has been doing so consistently for the the past fifteen to twenty years) purely on the back of free software.

Yeah, reminding people that corporations are not charities equals being a conspiracy theorist. What's next? Does saying that people breath air or that summer comes after spring make one a tinfoil hat wearer too?
/facepalm

[prodigy_]

That means any attempt to demand license fees for use of systemd would be met by forking the code.

Forks don't happen by magic. First of all, you need people to maintain a project of such scale. Do you keep up with the news? systemd is 550k lines already. It's horribly bloated but even if the first post-fork goal will be (as it should be) trimming all the fat it's still a lot of work. You can't just comment out everything you don't want. And this is especially true in case of systemd because the developers went out of their way to ensure it's a dependency for as much other stuff as possible.

It's extremely easy to get entangled in the systemd cobweb. But getting out is going to be very complicated and very time consuming.

[finalzone]

You know what the main difference between Red Hat monetization conspiracy theories and Canonical monetization conspiracy theories is? The fact that Red Hat is actually making money (and has been doing so consistently for the the past fifteen to twenty years) purely on the back of free software.

While doing stuff beyond GPL licenses by liberating most closed sources sources software they acquired.

[Hamish Wilson]

You know what the main difference between Red Hat monetization conspiracy theories and Canonical monetization conspiracy theories is? The fact that Red Hat is actually making money (and has been doing so consistently for the the past fifteen to twenty years) purely on the back of free software.

[Luke]

Systemd is LGPL-2.1+

Red Hat is behind systemd and Red Hat is very much like Microsoft - a heartless power hungry corporation. Their current intentions are obvious - they want a steel grip on the Linux userland and by now they pretty much have it. What their next step will be is less clear. They may take conservative approach and simply ask for a license fee but then again they could also come up with something more "creative". My bet is on the latter. Every systemd fanboy who is cheering today will cry tomorrow. Except the paid shills that is.

And "windowisation" means Linux - again thanks to systemd - becomes less and less of a *NIX and more and more of a crappy Windows-wannabe. Dbus and journald are classic examples of how we're losing control over the OS, of how things that made UNIX (and, by extension, Linux) great are thrown out of the window for no reason.

That means any attempt to demand license fees for use of systemd would be met by forking the code. The only way Red Hat could demand money would be to paywall the download site for systemd executables and limit source distribution to those either posessing the executable or paying for it. That would cause an instant fork, from either the last version distributed from an unrestricted site, or from single paid copies. One paid copy could brew a cut and paste redistro fork with hundreds of millions of installs, and under the terms of the LGPL Red Hat could do nothing about it. It would be almost like music filesharing after the musicians sold a copy under conditions requiring to be placed in the public domain, something they might do to get back at a predatory record company.