Suberian is an operating systems design project I've been working off and on for more than a decade now. Why bother with an OS project? Haven't the legions of morons proved the Unix / Windows duopoly cannot be broken? Haven't the (non-existent) legions of OS designers proved that Systems Software Research Is Irrelevant?
Well the problem with that view is that Unix and Windows blow chunks and the market is overripe, stinkingly so, for a revolution. The only reason it hasn't happened, aside from the distinct lack of vision is that everyone working on OSes is being a good little engineer or otherwise targeting engineers. IOW, nobody gives a fuck about ordinary users. One of the ways this plays out is that they work on and market some esoteric crap that users never actually care about. Instead of something users do see and do care about like, say, the UI.
And let's not forget about that lack of vision. Vision which can only be achieved by someone capable of both synthesis and analysis. Unfortunately, those neurocognitive traits are in rather extreme high demand so anyone displaying them both is going to get sucked up into and coopted by the capitalist system to do goodness knows what useless corporate shite for the rest of their lives. It's not like an operating system is a lightweight product either. It's extremely complex so for one person to design an entire operating system, that person's going to have to be at loose ends for years on end.
So okay, I've been working on this for over a decade. Why blog about it now? Why now of all times? Because it's nearing crunch time. The key parts of the design are all being finalized. I've seen the light at the end of the tunnel. There's only one major UI and one marketing issue left to figure out, the rest will be straight out implementation details. I'm not counting the sub-project I've mostly pushed back to Year 5 after initial release. Yes I do know the meaning of the word prioritizing, even if I've somehow managed to design almost everything in reverse order of implementation.
So that's Suberian. Right, I should actually describe it. Well, it adheres to every single high-level principle I could think of except Functional's Referential Transparency. There's not much point in listing everything including the kitchen sink. If there is to be a guiding principle above any other, it's Reflexion. I want it to be possible for Joe Schmoe (as opposed to Neo the H4ck0r) to be able to find his way to some random process, pause its execution, and fiddle with the stuff it has access to. And I want Joe Schmoe to understand what he's doing despite knowing fuck-all about programming. That's what reflexion is meant to be kiddies.
And that actually answers 'Why Suber?'. Because of Peter Suber's The Paradox of Self-Amendment and even more importantly his Nomic Game. Well, that and Suber is a pretty good last name.
Thursday, September 3, 2009
Subscribe to:
Posts (Atom)