I dont understand why u have to go about destroying hardware, Should'nt shutting down perimeter ingress and egress routers provide enough security that information is not going out of the location ?
And then cleanse internal systems in your own sweet time.
Theres definitely a difference between a rapid cycle throwaway crap vs I need this to work flawlessly for next 10 years stuff. Recently had the chance to troubleshoot an unconnect system which conked out, on the new jeeps and realized would gladly take an outdated stable interface vs bleeding edge stuff, especially when the only way to turn on a/c and controls is thru interface. Need beats fancy shit any day especially on 102F days:)
Arduino boards as many have suggested is a good start for hardware. http://arduino.cc/en/Main/Software
below is a good rule of thumb
you can do almost anything autonomously on the arduino board itself as long as you dont require it to number crunch.
Ex: Simple instructions like bump sensor registers send 90deg turn to some wheel servo you can keep it local to the bot
If your bot requires visual cognition (locate a ball on a plane) and associated number crunching, offload that video even offload that camera from the bot - stream to a laptop or bigger comp so you send only control signals to the arduino board.
So your bot would do something like below
1. Localized decision making (balance, servo speed, damping , breaking etc)
2. Overall decision making - bigger goal of the bot offloaded on laptop/Comp, control signal sent to compliment or override local controls.
A "Robotics" project sounds way to generic, A little more detail on your end goal would help you focus better.
If you want a premade solutions with all interfaces I'd start here
Lego Mind storms
If you want to try your hand at control algorithms without spending a penny I'd start here (sharp learning curve)
If you want to visually do something with your robot i'd start here, various boards and controls are included.
If you want a bit more advanced hardware I'd start here
For pure visual processing fun, this actually is rolled into ROS and DRC sim i believe
See i wish more companies were like Adobe, they can see a losing argument and innovate on the the next steps. If this was some other big company like IBM or M$ theyd have code cronies hidden away on the dark side of the moon still chipping away at it..
The method they are using s called as haar cascades postulated by viola jones. I have used the same with opencv for a bit now.
It's basically passing An image thru progressive classifiers to get a final weight of match.
Microsoft may have done the training for generating the classifiers but the method has been around for a bit.
"Decision tree".... Pfffft.
I agree with the guy who posted the xorg.conf above, i have a similiar setup where i have 2 monitors hooked to a single graphics card (single head)
have xinerama disabled, and have a config entry in XOrg.cong which says leftof / rightof the monitor identifiers.
This is enough to achieve what the poster requested. Maximize a window will maximize to one screen only.
and yeah if u did your job properly in a good school , 4 years does matter in the way u approach a problem.
not neccessarily coding skills or the best way to hack a one liner, the approach and bigger picture is as important or if not more