But it's a trade off. Long-lived TLS certificates have always had the cert revocation problem. OCSP stapling never took off, so in the end the consensus seems to have been to decrease expiry date. (Mostly fueled by Let's Encrypt / ACME).

Relying on expiration rather than explicit revocation of course also assumes (somewhat) accurately synchronized clocks which is never trivial in distributed systems. In practice it put's pressure on NTP, which itself is susceptible to all kinds of hairy security issue.

I like to think of the temporal aspect as a fail-open / fail-close balance. These centralized solutions favour the former, and that's why we see this resulting outage.

For this and related reasons, such as enforcing protocol upgrades, most smartcard systems end up permanently online.

Yep let's not learn from that incident and wait until is offline for like 2 weeks, and be assured that will happen.

The topic is an opener to discuss MitID, electronic ID's in general, the protocols behind them, what happens when they fail, privacy, societies reliance on them or something similar.

Without a trusted device or Recovery Key, Apple may impose a security delay (24 hours to several days) before allowing a password reset. Getting new SIM and re-authenticating our life will be pain.

If only it was a uint32

well, luckily, that's not how money is stored, but instead, they're transaction based. Aka, that number you have is a calculated value, not a stored, arbitrary value.

Except...perhaps the central bank's, where they could really just generate that money as an arbitrary value to lend out to other banks.

footnote: of course, your account balance is cached, so that it is not recalculated over and over again...

I remember hearing that Zimbabwe, during its period of hyperinflation, had problems because the databases for the banking system couldn't handle a time with $100 trillion banknotes, and ATMs didn't work because of overflow errors.

If only they had used int128. :)

Your other complaints: 100% agree, the whole thing is a privacy nightmare.

I wouldn't count on a post mortem of any value. They still refuse to explain how the system has been abused in the past. Regardless of how hard I try, I fail to understand how it has been abused after QR codes was added to ensure presence at the device you're trying to authenticate at. The system feels secure, but has been abused a number of times and we're almost never told how.

I really like the centralised system, it makes navigating society surprisingly easy when compared to say, Germany or the UK.

The difference is that I sort of trust the Swedish government, they've never really done anything to breach that trust - up to and including their handling of COVID (while controversial, they took the stance of individual liberty and a "collective responsibility" over mandatory top-down systems).

The UK in contrast has a much more heavy handed relationship with the population, up to and including incarcerating people for saying the phrase "we love bacon" at a construction site or typing the letter "n" on social media. It's a different context entirely.

Also, BankID, the central system is a definite weakness, but you can have a card/pin device that still works, and it does work on grapheneOS, though it will complain a bit if you don't have google services installed... which I find hilariously awful...

Having lived in Germany it's quite different, but I'd argue the centralized handling of the CPR is actually quite convenient and doesn't meaningfully impact privacy. In Germany every authority has its own ID for you anyway (my password manager has a category "Government Primary Keys" for this), however that means that you have to provide all your information from scratch to every authority. This would theoretically lead to more privacy if we lived in 1926, but now computers are ubiquitous and a rogue government (like Germany is close to electing) can just correlate these keys together. Relational databases have existed for decades and JOINS are cheap. Thanks to surveillance capitalism by now we have very sophisticated ways to deanonymize people, the government can just hire someone to do it.

So the privacy in Germany is most often inconvenience for the citizen paired with hardly any privacy gain from a potentially hostile government. At this point I think the better solution is to avoid electing hostile governments. To Denmarks credit, they're currently doing that better than many other European countries.