According to the story you just commented on they apparently care quite a lot about cross-platform support since they want people to help them with it.

Well, except that in this case the entire story is that they want help making it work great on non-Gnome platforms.

An API is not removed just because it's deprecated. It just means that you are discouraged to use that function in new code, and that it *might* be removed in a later version. This is not uncommon in minor versions and you typically wait for a major version until you actually remove them, to preserve ABI compaitibility.

Torvalds has said he might rewrite the kernel in VB one day. With .NET going open source (one bit at the time) this might not be too far away.

Didn't they fixed most of the nuisances with 8.1 and 8.1 Update?

Sounds reasonable. I don't think they are legally bound to keep that promise, but that they spell it out like that is a good thing. An interesting question that comes to mind is if the promise also covers modified code, it looks like the definition of covered code only covers code published by Microsoft. But still, better than nothing.

They work fine for me on Fvwm, both from 3.12 and 3.14. Must be something weird going on in your environment.

Maybe you should actually take a look at the GTK+ git history. There's a ton of work going into cross-platform support. Recently a lot of work was directed at Adwaita and make sure that it is always included and works great on all gdk backends, especially the win32 backend.

Then why are all major distros either in the process of switching to systemd, or have already done so? They must be complete morons switching to something that is just pain and suffering!

CoreOS is not designed to be used on just one or two machines. It's designed for huge clusters where one machine rebooting it not a problem. It allows them to implement more reliable updates. You either get it or you don't, and can quickly and easily roll back to the exact same bits you used before.

It's not like the journal format is some state secret. It's documented and there are already several journal parsers to choose from.

If the major distros all switch to Systemd, which looks likely; then that's one less thing that prevents people from switching to another distro. If you want to belive in some kind of conspiracy around Red Hat, then wouldn't they be more likely to just invent their own proprietary init system and make sure that no one else adopted it?

Of course stuff was broken. Stuff is always broken. Sysvinit was broken. Systemd is better but is probably also broken, one day it may as well be replaced by something else. The good part is, we are making progress. Over time the new stuff tend to be less broken than the stuff that came before it. You might tell yourself that nothing was broken before, but chances are it just happened to work well for you. Because if it wasn't broken, then that would be the first time in the history of the planet that a software project was infallible; and that would be amazing.

My understanding is that the Ubuntu community does not have the manpower that it takes to maintain universe, and Canonical is primarily only intrested in maintaining main and restricted. What they really should do is disable universe and multiverse by default.

The Ubuntu security team (which is mostly paid Canonical employees) provides security updates for packages in the main and restricted component. Packages in universe (such as owncloud) and multiverse are not supported by the security team.