PS: makes me wonder if we might eventually see something like "tokens not transferred since ${some generous date horizon} will get blacklisted starting ${some era switch date reasonably far into the near future for living address to create a lifesign transcribe}" merged into mainline. Or even "get reissued to miners". If that ever happened it would be at a time when total supply uncertainty would already noticeably influence price (basically: if you use it as a value store you need to hedge against possible old god address revival), and most buying/selling during the runup would be bets about the amount of dark addresses lighting up in time.

PPS: obviously I have no idea about governance of the code, but I do acknowledge that very few changes have ever been accepted (as evidenced by the high number of long running forks). I wouldn't expect any of this to happen before the known alive amount of tokens has become a tiny fraction of the dark parts.

Blacklists will be incredibly contentious and will likely never get merged. Even if they do - a fork is guaranteed to happen.

Undoubtedly. But my guess is that the threshold for blacklisting stale addresses would be noticeably lower than the threshold for any blacklisting based on, well, "governance".

You could even conjure up an argument based on block chain "balance of hashrate" security, which I presume is the best type of argument with that audience: "what if, after mining rewards have run out, everybody is just doing value store? No transaction fees for those selfless miners! (will they still be called miners?) One by one they will close their datacenters, difficulty goes down and suddenly the gates are wide open for 51%. Quick! Introduce mandatory keepalive ping transactions or we are doomed!"

Not that I expect this to happen, but I find these musings far more entertaining than the ups and downs of the market.

But surely most of the coins were lost when they were cheap rather than now? If that's the case most of the deflation stemming from lost coins already happened and isn't going to happen in the future, so it's not really a big issue.

Unless they were only presumed lost. Imagine someone had the power to suddenly activate whatever fraction of the contemporary USD supply their family had in 1800, proportionally scaled to the amount of USD circulating now. And in the deflatory bitcoin world, the amount of tokens in circulation might eventually end up being merely a tiny fraction of what some presumably dead addresses hold. An address like that lightning up would be as if the Fed suddenly announced that they just issued a multitude of the USD previously in circulation. Not to banks, as credit, but to some guy (unnamed). "Nice currency you have there, world economy, I have the other half of it. Let's be friends shall we?"

Of course these are more hypothetical end game states than immediate threats (I think, I have no idea what fraction could currently be considered dark), but we can't just ignore theoreticals like that.

Would they even be able to cash out though? Even a small transfer from a presumed-dead wallet can trigger a massive market crash in your scenario. Now if they just want to watch the world burn, that’s another matter.