I wonder when exactly Satoshi's coins began to be considered "lost" or "frozen".

I believe in Lopp's argument that quantum "recovery" is a theft to all Bitcoin users is completely false. As I already wrote, these coins are part of the circulating supply.

Let's ask ourselves some questions. Would the following scenarios be considered "a theft from everyone"?

1) Satoshi reappearing and spending or selling his coins.
2) Another early Bitcoin user (or their heirs, like in the case of Hal Finney) appearing and moving the Patoshi coins, claiming that these weren't mined by Satoshi but by themselves.
3) Hackers stealing the Bitcoins from some of the Patoshi addresses, because they were able to hack the private key from somewhere (e.g. an old computer or server).
4) Hackers stealing the Bitcoins from some of the Patoshi addresses, because the RNG which generated the keys had a flaw / low entropy.

I think at least the first two are not even worth a discussion: they are completely legitimate Bitcoin transactions like all others, and would end demonstrating that these coins were always part of the circulating supply.

But the third and fourth case are things that have happened with millions of other coins (and above all case 3 is happening literally every week or so). And while these hacks are of course theft, they are not considered to be a threat to Bitcoin's value or a "theft to everyone". Because their original owners could have sold them too.

The gap between case 4 and a quantum theft is minimal.

Thus, the whole BIP is based on a completely flawed concept of Bitcoin's economy.