3 ms·
I think his point though is that the wireless technology itself should make this impossible. The fact that it's not provably secure is problematic, because it's
by rickycook 11y ago
I think his point though is that the wireless technology itself should make this impossible. The fact that it's not provably secure is problematic, because it's all well and good to say "you're not allowed to" until someone subverts the rules. People break laws all the time, including police
- guelo 11y agoThere are very few violations of the FCC's rules in general partly because the FCC regulates and certifies hardware manufacturers of radio equipment, so it's very difficult to purchase equipment that doesn't follow the rules. You can see this happening in the Wifi space where the FCC is clamping down on dd-wrt at the device manufacturer level because the software allows people to violate that frequency's rules. People with the right skills can build custom equipment to violate the rules but it's rare. And for the few people that do try to do things like pirate radio the FCC hunts them down and the penalties are stiff. My point being that for the most part the FCC's regulatory scheme works at keeping people playing by the rules without extra layers of technically complicated security.
- jakeogh 11y agoIs it accurate to state they have made illegal (via 'rules') to use open source (user controlled) EM comm above 100um (below 3000GHz[1]) (except a few windows)? Or is it 9kHz to 275GHz? I bet when line of sight NIR comms (there's a nice window around 1550nm) get popular, they will (to stay relevant) want to 'regulate'[2] that too. Prior restraint is the problem here, anyone can mess with the spectrum with basic electronic skills, mandating 'approved' software seems less than pointless. [1] http://www.spectrumwiki.com/wp/allocations101.pdf http://www.spectrumwiki.com/wp/allocations101.pdf [2] https://www.youtube.com/watch?v=gbYXBJOFgeI https://www.youtube.com/watch?v=gbYXBJOFgeI