first of all, you can only broadcast "on top" of something if you're using an AM mode, and even then who wants to be talking through a spurious signal? Would you wanna have random voices or other sounds played over a phone call, interrupting conversation?
Why are you rationalizing / defending categorically anti-social behavior?
> Would you wanna have random voices or other sounds played over a phone call, interrupting conversation?
For the younger readers: this used to be a not uncommon thing in the POTS era, "lines crossed". And it was indeed annoying, especially if it persisted after hanging up and calling again.
Why would there be? Do you get random voices on a VoIP call done over wifi (radio) or there is interference? It sounds like your protocol needs more work.
> It's not illegal and is a core part of modern radio protocols.
LOL, yes it's for most practical purposes on hobby ham bands.
If you wanna be pedantic about it, the FCC rules for ham operations doesn't mention encryption, it mentions encoding instead [0]. So yes, if you encrypt or encode without obscuring the meaning (by publishing your encoding spec or encryption keys) then yes you can "encrypt". digital mode radio works that way.
Yes pretty much. There are a lot of digital modes for different use cases. Usually packet radio is referring to APRS. You can encode all you want as long as you make the encoding/decoding spec public and as long as the decoded information still comply with the rest of the rules like no commercial use, no profanity, call sign announcement, etc.
most equipment doesn't use trunked protocols like DMR, because it's vastly more expensive, and more difficult to use that honest-to-goodness analogue radio.
Plus, the hypothetical we're discussing here is that some non-ham asshole left a spurious transmitter running. Do you think that a trunked magically protects people from interference? No.
Finally, yes, encryption is illegal to use on ham radio.
Beauty? Please, there are messy things all over math, this appeal to beauty sounds like over-romanticized pearl clutching.
Ever seen the Chudnovsky algorithm for Pi? People are plenty capable of creating very important ugly math on their own. Not at all unique to LLMs, nor a valid critique. Do you understand the significance of proving that integer multiplication is simpler than O(n log n), even by a little bit? This is exciting! Who knows where this might go or what'll be learned from this
It's true that no benchmark communicates the whole picture, and we won't really know how it behaves until weights are out, but the performance here doesn't seem particularly groundbreaking just based on the benchmark.
One has to hope that this is mostly just the "uncanny valley" of AI as it climbs up the ladder of general intelligence.
At some point I think AI technology available to us will be advanced enough that "slop" is largely a thing of the past. Some of the issues here will still remain, but hopefully at that point a PR review could be largely automated, in a trustworthy way.
Right, the article does seem very, uh, doomery. The author has prematurely spelled the end of open source when we've hardly had time to contend with this.
Why are you rationalizing / defending categorically anti-social behavior?
reply