{ "version": "https:\/\/jsonfeed.org\/version\/1.1", "title": "Oleg Andreev: posts tagged deutsch", "_rss_description": "Software designer and crypto-anarchy historian. about • twitter • telegram", "_rss_language": "en", "_itunes_email": "oleganza@gmail.com", "_itunes_categories_xml": "", "_itunes_image": "https:\/\/oleganza.com\/pictures\/userpic\/userpic-square@2x.jpg?1686684482", "_itunes_explicit": "no", "home_page_url": "https:\/\/oleganza.com\/tags\/deutsch\/", "feed_url": "https:\/\/oleganza.com\/tags\/deutsch\/json\/", "icon": "https:\/\/oleganza.com\/pictures\/userpic\/userpic@2x.jpg?1686684482", "authors": [ { "name": "Oleg Andreev", "url": "https:\/\/oleganza.com\/", "avatar": "https:\/\/oleganza.com\/pictures\/userpic\/userpic@2x.jpg?1686684482" } ], "items": [ { "id": "47", "url": "https:\/\/oleganza.com\/all\/cryptographic-protocols-are-popperian-bitcoin-is-quantum\/", "title": "Cryptographic protocols are Popperian, Bitcoin is quantum", "content_html": "

While working on Flame<\/a> I have recently read both books by David Deutsch<\/a> and realized some of the ideas expressed there are explicitly baked into the
\ncryptography, Bitcoin and related applications.<\/p>\n

First, the idea that “we can’t ever prove knowledge, but can continue failing to disprove it”<\/b> is how pretty much all of practical cryptography is defined. Symmetric cyphers and hash functions are based on non-linear bit-mixing techniques that are pure art (just look at the diagrams from authors of Keccak!) with a combination of explanatory framework + statistical analysis that assures us that it’s unlikely to learn the input from the output. Every cryptographer knows that’s the best we could do until someone breaks it, like it was done with Enigma, then DES, then MD5 and SHA1. Asymmetric number-theoretic inventions such as RSA and elliptic curves have deeper explanations, but they are already broken in theory by yet-to-come quantum computers.<\/p>\n

Second, the idea that laws of physics should explicitly forbid certain outcomes<\/b> sounds very similar to how a well-engineered consensus system works, such as Bitcoin. In a communication protocol it’s pointless to define allowed messages because it’s up to users to choose what to transmit, but it’s all about preventing violation of invariants and restricting excessive use of resources. This goes even further when consensus system needs be upgraded without breaking backwards compatibility: new features are introduced by restricting the set of allowed messages, not by expanding what’s allowed. This way, network nodes that did not upgrade blindly accept new messages as valid per old rules, but newer nodes apply stricter checks, rejecting invalid messages. In Bitcoin it’s called “soft forks” and that’s how it has been functionally upgraded over years without risking parts of the network to get stuck.<\/p>\n

Third, Bitcoin exhibits both quantum and popperian logic<\/b> in its approach to transaction finality. The protocol does not define when any given transaction is considered final – that’s up to the recipient to decide when they make a decision to ship a product in exchange for receiving a transaction. This is because Bitcoin is not a single chain, but a tree of chains – each with its own cumulative proof-of-work. Nodes automatically switch to the valid chain with the heaviest proof-of-work and take the risk to accept payment when they feel like it’d be too expensive to revert it by making a stronger chain. At the same time, a weaker chain is never discarded by nodes and is always re-broadcasted – this is because any time in the future it might become stronger. This flexibility is what helps resisting all sorts of censorship attacks on the network – some nodes may buy temporarily on the weak chain, but eventually may get the most up-to-date data and join the majority. This is very popperian because no chain is ever fully final, just like our knowledge isn’t – we may always correct errors by inventing something new or by going back to an idea that we didn’t understand well enough. And it is sort of quantum because the network maintains multiple possible worlds and “decoherence” happens when people decide to act on some version of the “current world”.<\/p>\n

Fourth, the application of constructor theory to consensus<\/b>. Decentralized consensus systems are hard to scale because they are the opposite of “distributed systems” — instead of spreading the computation among nodes, every node bears the cost of verifying all the transactions made by all users. Gregory Maxwell, a renowned Bitcoin pioneer, always advocated that “blockchain is a court” – meaning that “smart contracts” must be executed by the network as a last resort to resolve the disputes and provide minimal “witness information” to prove the outcome. This is to minimize the cost of execution. This idea opposes “Newton’s laws of motion” that are often implemented in other blockchain networks, notably Ethereum: there the entire network is a simulation of a giant classical computer that explicitly computes every step as specified by the programs. Such explicitness hurts scaling, while Maxwell’s idea allows for infinite scaling, provided we have a clever enough proving system. I haven’t looked deep into the constructor theory yet, but I have a strong feeling that there are important parallels here: the blockchain network specifies the “laws of physics” that permit and forbid some transformations (“transactions”), while the “constructors” are the off-chain (off-consensus) entities – humans and their software that have to figure out how to satisfy certain constraints without bothering the rest of the network with it.<\/p>\n

I’ve read Deutsch’s exposition of all these ideas only recently, after living and breathing variants of them within the scope of applied cryptography, and that made me very excited. The chapter “significance of life” in The Fabric of Reality is simply jaw-dropping. I got excited about Bitcoin when I realized how it may shatter traditional hierarchy in civilization: asymmetric security (“harder to steal than to protect”), censorship resistant money that helps freedom of speech and experimentation, and a chance against large-scale destruction by wars due to de-concentrating economic powers. Seems like we just scratched the surface of solving the problem of mutual distrust (prisoners’ dilemma) by building systems that are so hard to manipulate, that an honest cooperative behavior is the only way forward. Maybe decentralized money is the way humans spend less time fighting each other and more time making progress?<\/p>\n", "date_published": "2026-11-10T11:00:48+00:00", "date_modified": "2026-11-10T13:04:27+00:00", "tags": [ "bitcoin", "deutsch", "popper" ], "_date_published_rfc2822": "Tue, 10 Nov 2026 11:00:48 +0000", "_rss_guid_is_permalink": "false", "_rss_guid": "47", "_rss_enclosures": [], "_e2_data": { "is_favourite": false, "links_required": [], "og_images": [] } } ], "_e2_version": 4171, "_e2_ua_string": "Aegea 11.4 (v4171e)" }