I apologize for that statement. I made it before I knew of 9Front and 9Atom. From what I saw, the code hadn't changed in a long time, and wouldn't boot in any environment I had. All that is false when you take into account 9Front and 9Atom. I now have 9Front running fine, and, in fact, I am renewing a port of an OO language extension to it.
Blake On Mon, Dec 16, 2013 at 3:26 PM, Oleksandr Iakovliev <yshu...@lynxline.com>wrote: > On 2013-12-15 18:05 , Blake McBride wrote: > > In spite of some really great ideas, I think we'd all agree that Plan-9 > has no real future. > > > I would not agree about that. If you would try to have a look at coming > future tendencies, you would be notified that there is coming what is now > named as "internet of things" where a lot of material objects in your > environment will have very small chips which would like to communicate to > each other and so on (there are already "scary" news about arrested china > transport containers of electric-irons and kettles which have some extra > less 1cm chip/device to listen for open wifi nets and spy on them or do > whatever they are programmed ;) ). > > Another tendency which is not so clear now but it is also coming: > computers/devices/systems/grids which perform actions being same time what > is called interface-less (good example is your car which automatically > identify you by sensors and wireless key/cellphone in your pocket when you > touch cardoor and then system just unlocks that - lot of computations, > communications and same time interfaceless). > > When you try to add these two tendencies to each other it would look like > that the next generation OSes should be much close to Plan9/Inferno because > it should easily cover connectivity and inter-communications of these grids > of tens/hundred/thousands of chips/soc/devices per 10 cubic meters around > you or worldwide (btw you can just read story about "bad bios" and suspect > of ultrasonic communications). They(OS) should be simple regarding internal > design. Parallel programming, computing/resource sharing, CSP, etc is > highly required and should not be complicated as it is now in world of > Unix/Linux/MS/Apple and should be possible to be programmed by individuals > or small groups. > Why not MS/Apple-like solution - because then such "nets" will be closed > and not really manageable at all. > Why not Linux - it is already over-sized and overcomplicated and highly > resistive to design changes, so even an admin with 1meter beard cannot see > all especialities of these such system/nets and cannot administering such > grids manually. Also consider the security of these complicated systems as > effect of simplicity of design of each part. > Regarding Apps - Plan9/Inferno have "reverse" idea: instead of App to > support environment where it has to run, it makes the environment to fit > the App needs - much more productive, stable, manageable. > It should be something simple, easy to join in swarm. Then interface part > does not have such huge value - even if it is ms system with browser - this > part does not play "key" role anymore. Plan9/Inferno or their derivatives > now have great chance for resurrection aka phoenix, but not as your laptop > OS with nicely drawn weather/news widgets or animated icons, though even > this is possible. > > just my 2cents for what we may see next >
