Shor's algorithm has been known for a while now (apparently since 1994) and is the main reason quantum-resistant cryptography became an important research subject. The article explains it nicely (for someone like me who doesn't know nearly enough physics or maths to fully understand the technical parts), but this bit at the end ruins it a bit:
> Rotate everything that lasts >10 years to pure PQC now
The author suggests switching to Post-Quantum Cryptography which uses relatively new ciphers that haven't been as battle-tested as older ones like RSA and ECC. Back when those were introduced, there weren't any stronger ciphers at the time, so if they were broken, at least people knew they did the best they could to protect their data.
Now, however, we have standardized encryption with (to the general public's knowledge at least) uncrackable algorithms (provided sane key lengths are chosen), so doing anything that could weaken our encryption makes us worse than the baseline. This proposal is theoretically stronger, but it is unknown whether it will stand the test of time, even with today's technology, due to it being relatively new and not widely deployed.
The standard practice of rolling out PQC is using it as an additional layer alongside current encryption standards. This adds redundancy, so that if one is broken the data will stay safe. Using only PQC or only RSA/ECC/whatever makes the system have a single point of failure.
FYI, this is exactly what governments want (I'll let you guess why). This related post was on the front page just a few days ago: https://news.ycombinator.com/item?id=46033151
> Rotate everything that lasts >10 years to pure PQC now
The author suggests switching to Post-Quantum Cryptography which uses relatively new ciphers that haven't been as battle-tested as older ones like RSA and ECC. Back when those were introduced, there weren't any stronger ciphers at the time, so if they were broken, at least people knew they did the best they could to protect their data.
Now, however, we have standardized encryption with (to the general public's knowledge at least) uncrackable algorithms (provided sane key lengths are chosen), so doing anything that could weaken our encryption makes us worse than the baseline. This proposal is theoretically stronger, but it is unknown whether it will stand the test of time, even with today's technology, due to it being relatively new and not widely deployed.
The standard practice of rolling out PQC is using it as an additional layer alongside current encryption standards. This adds redundancy, so that if one is broken the data will stay safe. Using only PQC or only RSA/ECC/whatever makes the system have a single point of failure.
FYI, this is exactly what governments want (I'll let you guess why). This related post was on the front page just a few days ago: https://news.ycombinator.com/item?id=46033151