I started with Yellow Dog on Apple PowerPC hardware, which I mostly used for servers, but played around a bit with various desktop environments. At that time I wasn't doing serious work so much as I just wanted to know as much about various platforms as I could. My primary OS for day-to-day computing was classic Mac OS (System 7, OS 8, OS 9) at that point.
As I moved to x86 hardware and Ubuntu became popular, I used Ubuntu as the primary OS on my development machine at work, and Debian on my servers. Nearly all of my servers (those that don't require anything bleeding-edge) still run Debian stable. Around the same time, I was running Mac OS X as my primary day-to-day OS on my personal machine.
Two or so years ago, I got tired of Ubuntu trying to push a particular desktop "experience" and having multiple layers of indirection and automation to deal with if I wanted to change anything. I briefly passed through Gentoo, which I enjoyed, but found that it required a little bit more time to maintain than was practical for me. I tried Arch and found it to be a good balance between customizability and niceties like binary packages. I switched to Arch on my main desktop at home, still using Mac OS X as my primary on my laptop.
A few months ago, I also switched to Arch on my laptop, more or less removing the Mac from my workflow.
If a high-level language like Python works for what he wants to do, then there isn't any reason to ever give it up. Code doesn't become magically better when it's written in C. Good, fast, and stable code for many applications can be written in high-level languages too.
If he goes on to become a proficient programmer, he will recognize that sometimes it is necessary to go to a lower-level language, whether it be for performance, linking against C libraries, or writing system code or plugins that must be directly executable. When that time comes, he'll learn C or something else. If he never gets to that point because he's satisfied staying at a higher level, then what's the problem? He'll have developed some great high-level programming skills he'd never have if he lost interest because the learning curve of a lower-level language was too steep.
The most important thing is that he learns stuff that keeps him interested. There is no perfect way to learn everything there is to know about computer science, so let him explore in the directions he wants to go. If he gets "corrupted" by a particular approach, then he probably didn't really have the interest to become a computer scientist in the first place.
Getting rid of the ground prong at the plug won't remove the circuit ground. The neutral prong is still ground in this sense. The ground prong is intended to be connected to the metal chassis, so that if a wire comes loose inside of an appliance and contacts the chassis, it will be shorted to ground instead of causing the chassis to go live.
The reason that there is an additional ground prong and the case isn't just connected to the neutral prong is that it's easier to mess up the wiring of line and neutral at the socket, or use an adapter that's not properly polarized, etc. It's harder to plug the ground prong into anything that's not ground.
If you cut off the ground prong, you're just removing this protection; the circuit ground is still on the neutral connector, so you're not protecting yourself from this attack.
There's a difference between ISPs, whom you are paying for access/bandwidth, and private Wi-Fi networks. The GP implies that people will be forced to open up their private networks, which is not the case and has nothing to do with network neutrality.
The whole point of having the sealed membrane keyboard design is that it's a lot harder for dirt to get into it. The whole top of the keyboard is a rubber membrane with no openings, so there's nowhere for dirt to get in. Also, unlike other membrane keypad designs, the membrane itself does not provide one of the contacts for the key. It merely applies pressure to the underlying plastic layers that actually have the contacts. The top plastic layer also has very few openings. It would be very difficult to get enough dirt into the keyboard through normal use (even in dirty conditions) to cause the keys to stick.
Mine developed a sticky control key after a few months. Opening up the laptop (compared to most laptops, it's easy to get into) and peeling back the top rubber membrane (it's lightly glued down to the plastic layers), rubbing the affected area to make sure the contacts were not sticking together, and reassembling the laptop seems to solve the issue; I haven't seen the control key stick since.
You could also make the circuit you describe look like a MP3 player or any other commercial product. Any circuit that you can build on a breadboard can also be constructed on a professional-looking PCB in a plastic enclosure. If you're going to argue that there could have been something malicious on a breadboard, by the same logic you'll need to ban nearly everything electronic.
Slashdot Mirror
I started with Yellow Dog on Apple PowerPC hardware, which I mostly used for servers, but played around a bit with various desktop environments. At that time I wasn't doing serious work so much as I just wanted to know as much about various platforms as I could. My primary OS for day-to-day computing was classic Mac OS (System 7, OS 8, OS 9) at that point.
As I moved to x86 hardware and Ubuntu became popular, I used Ubuntu as the primary OS on my development machine at work, and Debian on my servers. Nearly all of my servers (those that don't require anything bleeding-edge) still run Debian stable. Around the same time, I was running Mac OS X as my primary day-to-day OS on my personal machine.
Two or so years ago, I got tired of Ubuntu trying to push a particular desktop "experience" and having multiple layers of indirection and automation to deal with if I wanted to change anything. I briefly passed through Gentoo, which I enjoyed, but found that it required a little bit more time to maintain than was practical for me. I tried Arch and found it to be a good balance between customizability and niceties like binary packages. I switched to Arch on my main desktop at home, still using Mac OS X as my primary on my laptop.
A few months ago, I also switched to Arch on my laptop, more or less removing the Mac from my workflow.
If a high-level language like Python works for what he wants to do, then there isn't any reason to ever give it up. Code doesn't become magically better when it's written in C. Good, fast, and stable code for many applications can be written in high-level languages too.
If he goes on to become a proficient programmer, he will recognize that sometimes it is necessary to go to a lower-level language, whether it be for performance, linking against C libraries, or writing system code or plugins that must be directly executable. When that time comes, he'll learn C or something else. If he never gets to that point because he's satisfied staying at a higher level, then what's the problem? He'll have developed some great high-level programming skills he'd never have if he lost interest because the learning curve of a lower-level language was too steep.
The most important thing is that he learns stuff that keeps him interested. There is no perfect way to learn everything there is to know about computer science, so let him explore in the directions he wants to go. If he gets "corrupted" by a particular approach, then he probably didn't really have the interest to become a computer scientist in the first place.
Getting rid of the ground prong at the plug won't remove the circuit ground. The neutral prong is still ground in this sense. The ground prong is intended to be connected to the metal chassis, so that if a wire comes loose inside of an appliance and contacts the chassis, it will be shorted to ground instead of causing the chassis to go live.
The reason that there is an additional ground prong and the case isn't just connected to the neutral prong is that it's easier to mess up the wiring of line and neutral at the socket, or use an adapter that's not properly polarized, etc. It's harder to plug the ground prong into anything that's not ground.
If you cut off the ground prong, you're just removing this protection; the circuit ground is still on the neutral connector, so you're not protecting yourself from this attack.
There's a difference between ISPs, whom you are paying for access/bandwidth, and private Wi-Fi networks. The GP implies that people will be forced to open up their private networks, which is not the case and has nothing to do with network neutrality.
The whole point of having the sealed membrane keyboard design is that it's a lot harder for dirt to get into it. The whole top of the keyboard is a rubber membrane with no openings, so there's nowhere for dirt to get in. Also, unlike other membrane keypad designs, the membrane itself does not provide one of the contacts for the key. It merely applies pressure to the underlying plastic layers that actually have the contacts. The top plastic layer also has very few openings. It would be very difficult to get enough dirt into the keyboard through normal use (even in dirty conditions) to cause the keys to stick.
Mine developed a sticky control key after a few months. Opening up the laptop (compared to most laptops, it's easy to get into) and peeling back the top rubber membrane (it's lightly glued down to the plastic layers), rubbing the affected area to make sure the contacts were not sticking together, and reassembling the laptop seems to solve the issue; I haven't seen the control key stick since.
You could also make the circuit you describe look like a MP3 player or any other commercial product. Any circuit that you can build on a breadboard can also be constructed on a professional-looking PCB in a plastic enclosure. If you're going to argue that there could have been something malicious on a breadboard, by the same logic you'll need to ban nearly everything electronic.