It's always kind of cute to see ridiculous headlines roll through. Taking a look at the paper and a stream of thought about it. - Researchers are affiliated w

It's always kind of cute to see ridiculous headlines roll through. Taking a look at the paper and a stream of thought about it. - Researchers are affiliated with a university they aren't kooks - These are physicists. They speak a different language. That will make decoding this paper difficult. It may have useful insight, but it is obscured by the jargon of a different discipline. Typically I haven't see continuous diffeq in crypto. >.> - Essentially I believe this is a two-directional stream cipher, with prior clear?crypt?text from both sides feeding into the input. - I'm not sure what unbounded precisely is being meant here with their coupling function. - Almost zero crypto papers cited (I counted two, and they sounded fringe-y by title and were also published in physics journals). Several key aspects have been neglected by the authors: - Lack of citation of the current state of crypto- they betray nearly no understanding of the terminology of crypto nor present the contribution to the field in relation to the field. - No analysis of the information-theoretic leakage of the communication system. - No analysis of keyspaces, expected break time for chosen|known ciphertext, etc. My take on it is that it's a well-intentioned but largely useless paper by non-experts. If they took the time to understand the field, it might have some valuable insights. I would guess the coupling function as an idea results in a system design that leaks information like a sieve about both Alice and Bob. It might be a restatement of something like switching keys periodically, but I lack the vocab to interpret.