Man, there was a lot to read through there. Whew! Let me summarize exactly what's happening here. We have a divide between the "what we have now is good enough" crew, and the "lets try some more experimental features" crew. Both sides have valid points. But, sadly, I need to side with one of them. Having splash support in our initscripts is a good thing. We're not forcing anyone to use a bootsplash. In fact, we're doing just the opposite. We're facilitating it. We're saying "if you want to, go ahead". For those of you who DON'T use bootsplash, I humbly suggest you take a peek at rc.sysinit and look at all the stuff we support an no one uses. There's a lot of it. We support quite a lot of stuff in there because, frankly, there's no better place for it. Now here's the thing. This *could* be modular. We *could* do this in a separate dir of scripts that are executed way in the beginning of rc.sysinit. In fact, we could even get crazy and name that dir "/etc/rc0.d" or something. /me smirks. We maintain a very simplistic init style. This is a good thing. The problem with doing so is here - for some features, we need to put support in our scripts. We can't have the support in external files or directories, or anything like that - that's the kind of initsystem we explicitly DON'T have. So here's my suggestion for the naysayers - rewrite our initscripts to support a SysV init style. Do that and I will gladly maintain a second initscripts package. However, right now all you're doing is fighting against the "curse of Arch" - talking real work to death, and killing motivation. In summary: I'm for it.