Hi all, A [thread](https://groups.google.com/g/bitcoindev/c/7jkVS1K9WLo) was recently started on this list about cryptographic agility in Bitcoin. I believe there are important limitations around that idea, and wanted to offer my own perspective. It's more a philosophical consideration, and is not intended as an argument against (or for) any particular proposal today.
The idea is giving users/wallets the ability to choose the cryptographic primitives used to protect their own coins, from a set of available primitives that may change over time. I think this ignores how important it is what others do with their coins. If others' coins lose value, for example due to fears about them becoming vulnerable to theft with cryptographic breakthroughs, so do your own due to fungibility, regardless of how well protected they are. As an extreme thought-experiment, imagine that tomorrow a new cryptographic signature scheme FancySig is published, with all the features that Bitcoin relies on today: small signatures, fast to verify, presumed post-quantum, BIP32-like derivation, taproot-like tweaking, multisignatures, thresholds, ... Further assume that within the next year or two, voices (A) start appearing arguing that such a scheme should be adopted, because it's the silver bullet the ecosystem has been waiting for. At the same time, another camp (B) may appear cautioning against this, because the scheme is new, hasn't stood the test of time, and isn't well-analyzed. These two camps may find themselves in a fundamental disagreement: - Camp (A), fearing an EC-break (CRQC or otherwise), wouldn't just want FancySig as an option - they would want (feasible or not) that near-everyone migrates to it, because the prospect of millions of BTC in EC-vulnerable coins threatens their own hodling. - Camp (B), fearing the possibility that FancySig gets broken soon, possibly even [classically](https://www.quantamagazine.org/post-quantum-cryptography-scheme-is-cracked-on-a-laptop-20220824/), don't want to just not use FancySig; they would want near-nobody to migrate to it, because the prospect of a potential break of FancySig threatens their own hodling. This is of course an extreme scenario, and in reality the divergence between camps may be much less incompatible, but I still think it's a good demonstration of the fact that despite being a currency for enemies, Bitcoin essentially requires users to share trust assumptions in the cryptography, and its technology in general. A small note aside: you can argue that it is possible for people to store coins insecurely, like in an OP_TRUE script output, or with private keys that are made public. I don't think this possibility affects the point above, because it is not about what possibilities exist, but which approaches people are likely to / asked to adopt. Nobody is incentivized or encouraged to store coins in an OP_TRUE. It is also good to ask what amounts are relevant here, to which I do not know the answer. Possibly, the prospect of 100k BTC becoming vulnerable to theft won't threaten the value of the other coins, while the prospect of 10M BTC becoming vulnerable may. Depending on where your belief about this lies, you may end up with very different conclusions. Still, to me it is clear that some threshold exists above which I would be uncomfortable with seeing that many coins move into an untrustworthy scheme, even if they are not my own coins. This brings us to the question then how at all Bitcoin users can migrate to new cryptography, because we cannot assume that secp256k1 will last forever. And I think the answer is essentially that it requires the entire ecosystem to change their assumptions. This does not mean that adding a new opt-in cryptographic primitive is infeasible or a bad idea; it just means that adding FancySig as an option is changing the collective security assumption from "secp256k1 is secure" to "secp256k1 AND FancySig are secure" once FancySig gets adopted at scale, and the discussion about adding new primitives should be treated with the gravity that entails. And it means that disabling secp256k1 EC operations (or near-everyone migrating to FancySig, but I think that is unlikely) is the only way to change the collective security assumption from "secp256k1 AND FancySig are secure" to "FancySig is secure". Because of that, if CRQCs or other EC-breaks become reality, or just the fear about them becomes significant enough, I do believe that disabling EC opcodes will be necessary in any economically-relevant surviving chain, due to the millions of BTC that are unlikely to ever move. Note that I am not arguing that "Bitcoin" will, should, or even can do this, and I'm quite sure that the inherent confiscation required will make the result uninteresting to me personally. It may be that such disabling only happens in a fork that doesn't end up being called Bitcoin, or it may be that this happens only after a CRQC has been demonstrated to exist. Still, given a severe enough threat I think it is inevitable, and I think that this means we shouldn't worry too much about what happens in chains in which EC is not disabled while simultaneously EC is considered insecure: such chains will be worthless anyway. -- Pieter -- You received this message because you are subscribed to the Google Groups "Bitcoin Development Mailing List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/bitcoindev/THqOJuI_s5C8B9jkklN73BB_Hzb9SsiLM6BFp4zFP3zWQoRevKoLVspdwjwh8NxxYbXwv4v6ikpStguW-QEvef4WgBZ7AQDz00P0st91FMA%3D%40wuille.net.
