I wrote a post yesterday to the OpenSolaris Approachability community – whose job it is to make OpenSolaris as comfortable as pink-bunny slippers (my words, not theirs – the analogy here quickly breaks down, since pink bunny slippers aren’t practical, whereas the things the approability guys are doing are eminently practical). I’ve been using their Network Auto-Magic prototype on my laptop for the last while, and it’s pretty good stuff!

Anyway, in this case, my post was concerning the worrying idea that users new to OpenSolaris may take one look at the system, decide that it doesn’t behave exactly like whatever UNIX-like operating system they’ve been using already, and give up on it there and then. It also tries to address the now perennial question appearing on various OpenSolaris discussion lists “Why doesn’t OpenSolaris use foo instead of blah“. Now, my idea mightn’t be viable, so if anyone can add constructive criticism to my post, or a better solution to the problem I’m posing (if indeed it is a problem we should be worrying about) then I’d be interested in hearing about it.

In general, getting more people to use OpenSolaris can only be a good thing – so whether that’s by making it more approachable, easier to learn, easier to install, faster than X or whatever, I think it’s a worthy cause, and I’ll do what I can to help.

Update: 3rd Oct 2006 Cool. Sounds like the guys at Nexenta have implemented something similar by hacking the execve call – interesting solution!