My read on it is he's saying that KDE 3.5 isn't great on older hardware either, so what's the point?

I appreciate Martin's contributions to KDE, but when I read that blog post it seemed like he was taking the Trinity fork personally, especially due to one of this three imagined reasons for the fork being "Haters of “KDE 4″ technology". There are many other reasons why someone might want to keep KDE3 going apart from the three he gave.

Also: "Seeing commits to “twin” really hurts and our current version of KWin is just better in any area."

That post was from February, though, so there's little reason for us to continue to dissect it now. I just use it as an example of some of the unfortunate reactions developers have to forks of their code, and a demonstration that those deriding forks should not be taken seriously until after some investigation confirms the validity of their derision.

While udev can be built from the systemd sources, it still requires building dependencies that are NOT needed by udev.

This was silly and even though a patch was created specifically to address this, it was nak'd by the systemd developers.

Disclaimer: I've been a Gentoo developer for about 7 years now. I'm not directly working on the fork but I am testing and throwing in suggestions for using it on ARM devices where it doesn't make sense to have everything systemd provides, but the devices are more powerful and busybox's mdev is underkill.