Is this all we should expect from new technology
For the tech people. Is this it? Have we reached the max of computer technology innovation for the next 30-40 years. Think about it. Computers can only get so much faster. GPUs can only have so much memory. The software scenario hasnt changed much. Propeller heads use linux and everyone use Linux, the other +1/2% use mac.
If I werent so lazy and 'working' I wouldnt mind creating an operating system.
March 17th, 2007 10:34am
You would probably base it on linux though.
son of parnas
March 17th, 2007 10:39am
Yea, of course; way too many drivers to go do something on your own. That or some other operating system, FreeBSD, OpenBSD, etc.
March 17th, 2007 10:40am
So it would pretty much be just like everything else.
son of parnas
March 17th, 2007 10:42am
No it wouldnt. It would be based on the linux kernel. But you can come up with your own operating system tools, package manager, windowing system, filemanager, etc, etc.
No point in rewriting drivers when linux supports a billion of them. That is basically what the linux kernel is, a large driver manager. With that being said, you might also need the base tools; so I guess you might restrict the user from using those. and then have your own tools on top.
March 17th, 2007 10:50am
> your own operating system tools, package manager, windowing system, filemanager, etc, etc.
So basically the same as now, just botized for added flavor?
son of parnas
March 17th, 2007 11:04am
Bot, maybe you me, and Sharkfish should work together sometime on something low-level'ish, like for a driver and such. I'm just sensing a convergence of interests, to whatever extent.
Do you like the idea of low-level programming?
Just throwin' it out there.
BOT, why do you bother coming to this board for serious discussion? I thought you were on Blah.
Personally, I find it hard to take like 90% of what's on this board as serious. I think I'm gonna stop coming here because I don't want to be confused by whether someone is actually being serious here. harhar.
Sure Linux Or B. Actually, I am not into low-level programming (at least anymore), but there would be a great need for an operating system that could manage drivers better, a driver package manager of sorts.
I always on #concatenative (factor programming language) in freenode. I want to build an operating based on the factor language where the tools are developed in.
They have already built a window manager:
March 17th, 2007 12:01pm
It's not all but it's most of IT.
Look at the light bulb - it's been incandescent for 125 years. Some one will say what about LED or whatever Australia is mandating, but that *is* the point - a major innovation once every 125 years (and not even a paradigm shift) isn't exactly a growth industry.
Look at the kids who won Intel's Science Talent Prize:
There's no CS in the list although CS has far lower barriers to entry than an experiment on "microRNA repression".
So yeah, with the exception of AI, 75% of the IT technology you see today will be around in 30 years (though more mobile and embedded devices I'm sure). How many times can people reinvent a feature of LISP? And LISP is 45 years old itself.
But this is a good thing - your JnEE skills will still be in demand in 2040.
sour grape snowflake
March 17th, 2007 12:21pm
I'll get back with you later on that, Bot. I think I missed the whole Forth, stack-based revolution. Not sure about it.
I think I want to avoid Factor right now. Not because of the language but because I need to start on an app now rather than once I've learned that.
your call, it will still be here.
Wouldn't it be more interesting the other way around? Write your own kernel and run all the existing userland tools on top. If you're into this sort of thing, check out hurd or plan9.
The existing userland utilities compile down to the posix API of the *nix kernel (or something close to it). He would have make the kernel API posix or write his own compiler for that OS, and since he's probably only compiling programs written in Factor anyway(?)...