3 ms·
There's very little emergency communication that wouldn't benefit from at least cryptographic signatures; in many cases it's also important to be able to commun
by kragen 5y ago
There's very little emergency communication that wouldn't benefit from at least cryptographic signatures; in many cases it's also important to be able to communicate private data in emergencies. Not allowing encryption makes amateur radio of very little use in emergencies.
There are lots of "telecommand" use cases that aren't to satellites, too.
The broader problem is that prohibiting new technologies in ham radio until they've been demonstrated to be useful (in your case, demonstrated to be useful in ham radio, despite being prohibited at the time, but most people are less blatantly unreasonable) is a recipe for, at best, chasing the taillights of commercial radio systems, and more commonly total technological stagnation and gradual abandonment.
- VLM 5y ago> Not allowing encryption makes amateur radio of very little use in emergencies. I can't even imagine what would require encryption ... maybe PII medical traffic under some peculiar artificial scenario and most pessimistic possible HIPPA interpretation? Now remember, that the stereotype of ham radio emcom is people roll up with no idea how to program their HT to the correct frequencies so they can't even participate. So demanding everyone participate in a GPG WoT or join a Kerberos domain to share encryption keys is an absolute non-starter. If the general public cannot participate without extensive preparation before the event, it means its not part 97 emcom, its just a poor low budget imitation of part 90, its not part 97 operation at all. Why not get GMRS licenses? Remember there is the perpetual battle in emcom between the people who think "provide emergency communications" applies to Big Brother despite his infinite budget outclassing any possible Ham response, vs the people who think emergency communications is for the people or the general public or even for the individual. Both sides like to denigrate the other side, unfortunately. So consider I was listening to a rando ham "involuntary storm chaser" report a tornado sighting to the NWS a couple weeks ago. How, in that scenario does encryption help? By making it impossible for the ham to talk to the NWS station because they never shared keys in advance or they shared the 2020 keys and not the 2021 keys? By making it impossible for the general public to hear the tornado report they would be safer not knowing? Consider something topical, like the hurricane in NO. A rando member of the general ham radio public reports hurricane related flooding. The value encryption provides is making it impossible for the ham radio operator to talk to people to make the report, and encryption also makes it impossible for the general ham radio public to help. "Real ham radio emcom" situation he'd just dial up the repeater and talk to net control and make his report and move on with life. Somehow encryption would help, but ... how? Edited to add: > There are lots of "telecommand" use cases that aren't to satellites, too. I forgot to agree with you on that. Repeater control links and the like.
- kragen 5y agoMaybe "very little use in emergencies" was an overstatement on my part. Clearly there is some important communication in emergencies that doesn't need crypto. But I think the vast bulk of communications necessary in an emergency is between people who need to talk to their family members to find out where they are, what condition they're in, and what resources they have. That information often needs to remain private between the people who are communicating. Right now it's not mostly sent over ham radio because it can't be. In fact, it isn't transmitted at all, so people die. Advances like Codec 2 dramatically expand our possibilities for communications in emergencies, but mostly for doing new things like that, things we're not already doing. As another example, suppose that, instead of one person reporting hurricane-related flooding you had a water-level sensor every 100 meters in a rough hexagonal grid, each transmitting a packet every 6 minutes carrying a 16-bit ID and an 8-bit water level. In theory this is an average of about 8 bits per second per square kilometer, so it wouldn't have to cause much interference or use much average power, but it's hard to get it to work under Part 15 rules because Part 15 doesn't allow you to average your power usage over 6 minutes, so you pretty much have to build a mesh network to get adequate range at the allowed power levels. You probably need to cryptographically sign the packets for the system to be reasonably reliable. (They don't need to be secret, but you need to be able to distinguish legitimate reports from griefers trying to divert emergency resources when there's no emergency.) Nowadays you might be able to do that with LoRa. But you could have done it 20 years ago on the amateur bands --- except that it's illegal. So again hams are chasing the taillights of commercial radio, the opposite of how it's supposed to be.
- sokoloff 5y agoIn an emergency, I can’t imagine having an amateur radio link available, having a need to comm with family, and deciding not to because the connection was unencrypted and unsigned. Airplanes are controlled by in the clear, unsigned radio comms every day. I don’t see why emergency comms over AR is a more compelling case for a secure link.
- kragen 5y ago