“Bitcoin is always and forever only probabilistically consistent, thus never CP [but always AP in the CAP]. This blog is incorrect, although it does illustrate a point about trading off availability latency for [a naively computed? c.f. below…] probability of consistency. Unfortunately the blog does not even mention probability.”
Bitcoin’s partition tolerance doesn’t mean partitioning is nice. Imagine aliens mining Bitcoin on Mars exploiting some technology for nearly free energy, at a million times the hash rate of on Earth.
[1] For n00bs, the Nakamoto consensus protocol rule is that the partition (i.e. fork) with the longest chain (actually to be more precise the highest cumulative proof-of-work difficulty) always wins and the other fork is orphaned.
Competing forks are resolved probabilistically w\high probability (by orphaning all but one) typically within up to ~6 blocks; but not if network partitioned such that the forks aren’t aware of the other (until much later).
This is an example of lying w\Probability & Statistics (https://en.wikipedia.org/wiki/Lies,_damned_lies,_and_statistics)— why I emphasize the true independent random variable is often the unaccounted (even unknowable) factor that can lead to such egregiously incorrect computations. The long-tail distribution (aka Black Swan event) is essentially failure of assumptions about the independent variable due to appearance of a Black Swan outlier. Accumulated inertia of such assumptions is the antithesis of Taleb’s antifragility.
But the Poisson process (i.e. a probabilistic) model is unaware of a partition running (e.g. on Mars by aliens) at a different hashrate. The “independent” random variable in the Poisson model is actually dependent on the probability distribution of partition events (i.e. an orthogonal probabilistic model)— something you will never see mentioned in any mathematical discussion of Bitcoin. Heck this isn’t merely theoretical nor so far-fetched given the radio signal propagation delay between Mars and Earth ranges from 4.3 to 21 minutes, making Bitcoin entirely unsuitable for an interstellar mined blockchain.
Hypothetically even brute-force, nearly cost-free, long-range attacks (analogous to a withheld/uncontactable partition) might someday be viable on Bitcoin employing hypothetically[2] a quantum computer employing Grover’s algorithm for preimage search[3] might be able to orphan the entire Bitcoin history since inception, c.f. *§5 Resistance to quantum computations* of Iota’s Tangle whitepaper[4] although quantum computers are no where near practical yet[3][5] and Daniel J. Bernstein showed that some (at least one) quantum algorithms (e.g. hash collisions) will **never** be more cost-efficient with a quantum computer.[3][6]
Note this posited attack presumes the profit/motivation from/for orphaning the non-quantum history subsumes appending quantum computing mined blocks to said history.
“Ethereum’s PoS implementation has been teased for the better half of a decade now, but with the beacon chain running for 18 months straight and successful live merges on multiple testnets the initial implementation is largely finalized.”
Dubious whether bugs and corner cases have been exorcised given the high likelihood of the natural oligarchy enforcing a preferred order:
“A supermajority (2/3rds) of validators is required to finalize a block, in case of a 50-50 network partition blocks would stop being finalized and attestation rewards would stop. Non-participating validators would slowly leak stake through the inactivity leak until online validators once again had a supermajority. This is the ‘self-healing’ mechanism that allows both safety and liveness.”
THIS ENTIRELY LACKS THE PARTITION TOLERANCE OF THE CAP THEOREM.[1] This design bleeds out the partitioned stake from every partition, so they are permanently forked off from each other unlike for proof-of-work where all but one of the partitions will eventually be orphaned when the network returns to normal.
Also this presents either an inferior liveness situation to proof-of-work or means for the natural oligarchy (probably insidiously, surreptitiously) extract value:
“Equivocation is punished by slashing up to the validator’s entire stake, so the attacker must commit to destroying at least one-third of all staked ETH. The cost to reorg a finalized block is several billion dollars, even at today’s depressed prices.”
Not that they have any incentive to do so when the natural oligarchy (w\majority stake) can more easily and effectively extract value insidiously instead. There are theoretical long-range attacks for said natural oligarchy in proof-of-stake against lock-up deposit penalty mechanisms, but this simply isn’t the preferred way that proof-of-stake systems mine the FOMO n00bs instead by surreptitious value extraction.
“One key difference is that the honest validators would have to explicitly band together to recognize one another’s attestations and override the fork choice rule, but other than that they can form their own child chain and the malicious supermajority would slowly bleed stake out of the validator set until the honest subminority has once again regained a supermajority.”
The majority stake chain can censor blocks and attestations! The bleeding based on inactivity of attestations is a double-edged sword that can be (even insidiously) turned against the subminority by the malevolent supermajority. The “he said, she said” dilemma of weak subjectivity[2] applies if instead the honest minority band together to create their own fork to bleed out what they claim is the dishonest majority.
In addition to some additional points I made at the aforelinked, Lyn Alden also refuted (https://www.youtube.com/watch?v=1m12zgJ42dI) the following linked section. Essentially she argued against the highest economies-of-scale have the most proof-of-work profit, because she alleges there’s too much risk (e.g. jurisdictional) and micromanaged opportunity-cost (e.g. making deals in situations where byproduct energy or natural sources are normally discarded/unharvested).
“The first time a node comes online, it must have some frame of reference to know how to bootstrap itself. This is not a problem unique to PoS; even a bitcoin full node needs to know which client software is valid, which IPs to bootstrap its history from, etc. PoS adds only minor additional trust assumptions here.”
Disingenuous hyperbole at least w.r.t. practical usability if not also the fundamental invariant.
Trusting your software client to be running the correct protocol is not a subtle distinction from the-human-user instructing the software client which of the SUBJECTIVE conflicting forks is the canonical one. For one because most users will be incapable of doing that without being vulnerable to deception (e.g. phishing attacks, etc).
Fundamentally as you know, there’s no way to ever OBJECTIVELY know for sure what was/is with proof-of-stake (or any non-proof-of-work) consensus system. That is so due to the possibility of a network outages/partitions, even if the user thought their software client was active in real-time for the entire history of the blockchain.
Either of the aforementioned mishaps with proof-of-stake only need to happen once for great harm; whereas, there’s no choosing of IPs with Nakamoto consensus, as the user’s software client can AUTOMATICALLY compare the results from innumerable IPs that claim to connect to the canonical gossip network, OBJECTIVELY choosing the longest (or more accurately the highest cumulative mining difficulty) chain. The exhaustive search for objectively can continue indefinitely, thus highly improbable the software client is going to be remain walled off from the objective reality.
“That's dangerous to use social consensus to ‘slowly bleed stake out of the [censoring] validator set.’ What if the censoring validator set is doing so because of government mandates…”
Indeed. Or for example if the natural oligarchy censors (i.e. ignores) the minority attestations (making them appear to have never been sent bcz not recorded) thus causing those honest validators to bleed out (due to recorded inactivity) even though they’re not malevolent and were sending attestations across the network. Thus further centralizing the stake (i.e. wealth) of the blockchain in the natural oligarchy. An example of why I had coined/adopted the term ‘Consortium’ blockchains for proof-of-stake:
Ethereum’s proof-of-stake (PoS) security theater (e.g. slashing, etc) presumes we’re n00bs unaware of the existence of the natural oligarchy coupled with the fact that “reality” in networks is a blame game (which is precisely what consensus is for, duh), and the majority decides who to blame and thus can paint any reality they want if they can agree to (which is trivial to lie about given the natural oligarchy is the majority). C.f. my references to ‘blame’:
Any technobabble minutiae counter argument is refuted w\the inviolable fundamental generative essence fact that the natural oligarchy (i.e. w\majority stake) has total control and change the reality as recorded by the protocol in all “proof-of-stake” (i.e. any not proof-of-work (PoW)) consensus systems (even those w\obfuscating names/concepts that all actually distill to natural oligarchy control, e.g. proof-of-activity, proof-of-importance, proof-of-elapsed-time, proof-of-contribution, proof-of-coverage, etc). Their designers/protagonists are unable/unmotivated to pinpoint a specific mechanism of the vulnerability, it always exists due to said insoluble invariant in non-PoW distributed consensus.
The argument that PoW is also vulnerable to a 50+% (actually less than 50% due to orphan rate) takeover is not in every facet an equivalent vulnerability because expended work, ASIC depreciation and sunk cost are not conserved unlike stake which is always conserved (for the natural oligarchy which can obviate all PoS protocol penalties in their favor).
For example, unlike in PoS, the natural PoW mining oligarchy can’t SUBJECTIVELY vacate liveness w/o forsaking revenue, ASIC depreciation and/or sunk cost. Alternatively can mine empty blocks and (via 50+% attack) orphan all non-empty blocks but this attack is OBJECTIVELY distinguishable from a flaky/partitioned network. Honest miners could choose to objectively fork away by having an automated Schelling point protocol rule to ignore empty blocks. Unfortunately Bitcoin’s PoW is flawed in that the miner’s block could be filled with transactions (txns) to self, paying all the transaction fees to self, which is surreptitiously as deleterious as an empty block. I posited an idea for possible improvement:
Note a 50+% attack orphaning all minority block (regardless if empty or not) does unfortunately progress subjectively to 100% centralization of mining, not objectively distinguishable from a flaky/partitioned network (ditto as aforementioned for PoS). This is the selfish-mining attack in extremis in that 100% of the rest of the network’s blocks are orphaned.
However unlike for PoS, there exists (except for CPU-only mining such as Monero’s RandomX) the nuclear *THREAT* (thus likely never necessary!) of hard-forking to change the PoW algorithm to burn all of the attacking mining oligarchy’s ASIC sunk cost instantly turning those ASICs into door stops. What is the equivalent threat in PoS? Burning the oligarchy’s stake will not work. In PoW, the natural mining oligarchy isn’t necessarily the natural token oligarchy. Whereas, in PoS the two are conflated. I wrote:
“Intrepid Vitalik’s naivete belies the fact that the ~80 - 99% are the power vacuum—and ‘nature abhors a vacuum and for vacuums of power it’s doubly true’—which are naturally harvested by the ~1 - 20% per the inviolable natural order of the power-law distribution of wealth and fungible resources. Any blockchain which burns ~80 - 99% of its stake (i.e. chops off its face, disembowels itself) to allow its ~1 - 20% economic minority to transact (i.e. to save its skin — the ~80 - 99% majority of thus nearly worthless human protoplasm) will isolate itself from anything economic¹⁵ (i.e. be worthless).
Thus not only would the oligarchy retain economic control on their fork, if there was some inane ‘social consensus’ witch hunt attempt to blame the merely suspicious (but not objectively provable—only subjectively—culpable) validators.”
Said nuclear threat isn’t entirely vacuous bcz the hard fork although potentially initially of much lower security (e.g. if no ASICs yet exist for it, though the economic pressure to create them pronto could be immense), is an air-drop wherein everyone can vote by which chain they sell to buy the other w\proceeds.
Note the block-censoring, PoW natural-mining-oligarchy would decrease the systemic hashrate by that of the censored minority (i.e. up to ~50%) thus presumably decreasing the price but potentially recoverable by the reduced competition and mining difficulty. Yet it’s unlikely this would not be perceived as an attack, thus cratering the price. Ditto for PoS, except that the damage (e.g. as compared to hashrate-driven valuation and security in PoS) reversible (e.g. potentially after profiting on shorting) without any forsaken ASIC depreciation and minority miners destruction (i.e. which do not return). More likely in both PoW and PoS is insidious trickle “selfish-mining” (as a catch all term for block censoring) to gain market share over the long-term and other forms of insidious value extraction[1] such as MEV (with objective randomness being a complete farce on PoS blockchains as I explained in my blog).
Jurisdictional arbitrage likely obviates governments successfully regulating PoS validators or PoW miners unless in some future NWO world government w\the Antichrist as the natural oligarchy. Meanwhile, ownership can be severed from control with appropriate exigent contingency plans for when an owner is arrested, e.g. distributed multi-sig and geographically distributed ASICs.
I’m more worried about insidious censoring of txns aligned to the interests of the natural oligarchy. My aforelinked posited improvement to Bitcoin (and any applicable PoS designs) would make including any txns (i.e. non-empty blocks) an all-of-nothing affair and/or make implausible for miners/validators/block-builders to know the txns before adding them to a “finalized” block. E.g., a superfluous txn and a required supposedly random number in each txn might defeat block builders’ a priori demand for txn deets bcz said number could be the hash of a txn to be revealed later— tho builders might block based on superfluous correlations.
In summary but c.f. my follow-up (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56467747), excluding a separate issue that PoS generates no production cost monetary value and can’t distribute tokens in a decentralized meritorious manner, and assuming the existence of the natural oligarchy, there’s little practical difference in effect between PoW and PoS on the operation and outcomes (although PoS has a severe liveness inferiority if there’s no a natural oligarchy), excepting that the PoS natural oligarchy has more cost-free control (i.e. rope-to-hang-itself-with) thus could f-up, (perhaps irrelevant if the natural oligarchy is preventing all conflicting forks) PoS’s weak subjectivity and lack of non-manipulated public randomness on PoS (i.e. a surreptitious insidious extraction of value[1]):
This at least in the near-term probably doesn’t manifest as any incentives incompatible outcome that destroys the value of respective blockchains. When the natural token oligarchy’s interests align with some global enslavement paradigm (e.g. the biblical Antichrist), blockchains will be lights out for human civilization.
[1] Jason Drake admits PoS stakers have huge opportunity cost (https://youtu.be/1m12zgJ42dI?t=5340), thus logically they must surreptitiously extract more value! Largest PoW miners hedging risk reduces their rate of market share accumulation w.r.t. smaller PoW miners who take more risk!
Lyn Alden *might* be correct that proof-of-work mining is not 50+% centralized due to the aforementioned current risks (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56417630). But unfortunately under an Antichrist NWO world government, those risks might eventually be vacated by the said supreme power, thus enabling mining centralization. The said nuclear threat might become impotent because the world would be too dependent on Bitcoin by then. Which is exactly the Biblical end game and what I was alluding to with my 2013 Bitcoin: The Digital Kill Switch.
Meanwhile, as Jeff Booth eloquently explains to Green Capital (https://youtu.be/xf4RqFuru2g?t=2913), Bitcoin is the only reliable, non-manipulable, “neutral” (https://youtu.be/xf4RqFuru2g?t=528) unit-of-account “measurement” (https://youtu.be/xf4RqFuru2g?t=761) of complete reset of the global order to one of massive abundance which invalidating (https://youtu.be/xf4RqFuru2g?t=643) the economic viability of the extant political systems (https://youtu.be/xf4RqFuru2g?t=186). Jeff cites (https://youtu.be/xf4RqFuru2g?t=1184) the blockchain trilemma (security, decentralization, scalability) as the reason proof-of-stake can’t have the aforementioned qualities. He first implies that Ethereum is insecure (which it has been but that was due to Turing-complete programmability: http://trilema.com/2016/to-the-dao-and-the-ethereum-community-fuck-you/). I’d argue proof-of-stake is “insecure” because it’s centralized (which Jeff also thinks), e.g. how I explained in this blog that the natural oligarchy can manipulate randomness to manipulate the outcome of the programmable smart contracts. Jeff states, “who would build a decentralized finance on top of a centralized, expensive database, it’s so ludicrous…they don’t have utility…I can’t see one thing that stays valuable in the crypto ecosystem [other than Bitcoin]…create my own coin and pretend it has value…they lobby governments with the money they’re stealing from people…they’re all going to go to zero anyways.” I also predicted that where I wrote in early 2019 (https://steemit.com/blockchain-scaling/@anonymint/lightning-networks-must-fail-if-it-succeeds):
“…(aka block size limit) has to be immutable otherwise the ‘social consensus’ involved to change it would (as democracy and voting always does) steal, squander, and destroy all the value…envision scenarios such as by politicians representing competing forks offering some benefits (http://esr.ibiblio.org/?p=984) …”
Jeff is failing to incorporate into his futurist, utopian model the inviolable invariants that the power-law distribution dominates fungible resources and that power-vacuums can only be sustainably captured by the most ruthless.
Assuming the existence of a natural oligarchy (which is mostly certainly the case, at least and if only for proof-of-stake), thus overly dramatized/emphasized (aka security theater) concerns about security (c.f. Power Ranger Jason Drake’s pointless and witless nonsense where he conflates security with non-out-of-thin-air monetary value creation: https://youtu.be/1m12zgJ42dI?t=5619) are vacuous/pointless given the natural oligarchy has no incentive to destroy the value of its own blockchain and instead an incentive to insidiously/surreptitiously extract value from it, which is much easier in proof-of-stake. Proof-of-stake exists to fool n00bs and mine the FOMO for value extraction. Because in neither case for Bitcoin or Ethereum, is either at any risk of being 50+% attacked until perhaps FOMO is dead and proof-of-stake has thus reached its utility expiration date. Upstart, fledgling smaller proof-of-work blockchains are more vulnerable to proof-of-work difficulty non-readjustment bomb attacks than proof-of-stake (http://trilema.com/2014/the-woes-of-altcoin-or-why-there-is-no-such-thing-as-cryptocurrencies/), but that’s irrelevant to the salient consideration herein.
Jason’s arguments about the value of the social layer are also vacuous:
Hilarious that Jason Drake argues (https://youtu.be/1m12zgJ42dI?t=4488) that reducing the liquidity of tokens originally created out-of-thin-air[2] in proof-of-stake generates more token value than expending them on external resources in proof-of-work to set the production cost valuation of digital commodity money. It’s no less wasteful than proof-of-work, while not imparting any cost-of-production valuation to digital commodity money:
“Something I've noticed with many developers and engineers (as I used to be an engineer) is that they are unable to see the bigger picture. They become so lost in the specific technicalities of a product that their view on macroeconomics is often extremely poor. This is common among most Ethereum advocates. They're all technically very savvy but their knowledge on economics is poor.”
Nonsense. USAF in 2017 was an attack on the Nash equilibrium thus will not sustain as Nakamoto consensus is only stable in said equilibrium. Specifically the impostor soft-fork created a (by now) ~17+ million BTC booty which means miners no longer have a single best strategy as required by Nash equilibrium. The novel pay-to-script-hash protocol is ANYONECANSPEND (in Satoshi’s legacy protocol) and will be completely fleeced (i.e. taken from wallets w/o touching the private keys because ANYONECANSPEND means a signature w\private key isn’t required to spend it) bankrupting for example BlackRock and all its corporate Wallstreet (even now the Wisconsin government and of course Michael Saylor who I tried to warn via his business mgr but was ignored) minions and sycophants (perhaps at the halving in 2028). The game theory and economics of the situation are inarguable, if you actually are not clueless a.f..
Contrary to Justin’s snowflakes delusion (https://youtu.be/1m12zgJ42dI?t=975), the social layer of proof-of-stake doesn’t override the natural oligarchy in any economic sense. I wrote:
“Intrepid Vitalik’s naivete belies the fact that the ~80 - 99% are the power vacuum—and ‘nature abhors a vacuum and for vacuums of power it’s doubly true’—which are naturally harvested by the ~1 - 20% per the inviolable natural order of the power-law distribution of wealth and fungible resources. Any blockchain which burns ~80 - 99% of its stake (i.e. chops off its face, disembowels itself) to allow its ~1 - 20% economic minority to transact (i.e. to save its skin — the ~80 - 99% majority of thus nearly worthless human protoplasm) will isolate itself from anything economic¹⁵ (i.e. be worthless).
Thus not only would the oligarchy retain economic control on their fork, if there was some inane ‘social consensus’ witch hunt attempt to blame the merely suspicious (but not objectively provable—only subjectively—culpable) validators.”
Even the OP blog author seems to be aware of this economic reality:
You’re ostensibly writing about topics outside your capability to fully grok the implications of what is actually going on. You should be more circumspect.
ʀᴀɴᴅᴀᴏ can be entirely hijacked by the natural oligarchy (scroll down to the discussion of proof-of-stake):
ʀᴀɴᴅᴀᴏ has no complete formalization and is heuristic of unknown vulnerabilities, unlike for example Dfinity’s (or Ourboros’) beacons which are formalized (but still suck in terms of liveness and fail to the natural oligarchy).
Proof-of-stake is dogshit and I will make this argument more comprehensively in other follow-up comments on your very myopic blog.
Note you did correctly acknowledge the inviolable existence of a natural oligarchy:
My (censored) comment on the CAP Theorem (https://en.wikipedia.org/wiki/CAP_theorem) distributed systems limitation:
https://www.goland.org/blockchain_and_cap/comment-page-1/#comment-2681576
https://www.goland.org/blockchain_and_cap/#comment-2681576
“Bitcoin is always and forever only probabilistically consistent, thus never CP [but always AP in the CAP]. This blog is incorrect, although it does illustrate a point about trading off availability latency for [a naively computed? c.f. below…] probability of consistency. Unfortunately the blog does not even mention probability.”
Bitcoin’s partition tolerance doesn’t mean partitioning is nice. Imagine aliens mining Bitcoin on Mars exploiting some technology for nearly free energy, at a million times the hash rate of on Earth.
Later when they broadcast their chain to Earth, all earthly Bitcoin wealth will be erased and fleeced (https://paulkernfeld.com/2016/01/15/bitcoin-cap-theorem.html#big-partitions).[1] Thus Bitcoin is only ever probabilistically consistent.[7] Bitcoin wealth is not absolute.
[1] For n00bs, the Nakamoto consensus protocol rule is that the partition (i.e. fork) with the longest chain (actually to be more precise the highest cumulative proof-of-work difficulty) always wins and the other fork is orphaned.
Competing forks are resolved probabilistically w\high probability (by orphaning all but one) typically within up to ~6 blocks; but not if network partitioned such that the forks aren’t aware of the other (until much later).
This is an example of lying w\Probability & Statistics (https://en.wikipedia.org/wiki/Lies,_damned_lies,_and_statistics)— why I emphasize the true independent random variable is often the unaccounted (even unknowable) factor that can lead to such egregiously incorrect computations. The long-tail distribution (aka Black Swan event) is essentially failure of assumptions about the independent variable due to appearance of a Black Swan outlier. Accumulated inertia of such assumptions is the antithesis of Taleb’s antifragility.
[Technobabble minutiae: probability of finality (https://math.stackexchange.com/questions/2356763/the-probability-behind-bitcoin) (i.e. consistency, c.f. also (https://bitcoil.co.il/Doublespend.pdf#page=7)) in Nakamoto consensus is normally calculated (https://ems.press/content/serial-article-files/11512#page=2) (actually approximated (https://arxiv.org/pdf/1801.07447#section.5)) as a (continuous (https://archive.ph/https://suhailsaqan.medium.com/explaining-bitcoin-mining-as-a-poisson-distribution-92b2481fb80f)) Poisson process with an exponential distribution of time between blocks (https://bitcoin.stackexchange.com/questions/25293/probablity-distribution-of-mining) and the (discrete) Poisson distribution (https://en.wikipedia.org/wiki/Poisson_distribution) for blocks per period, wherein the average constant rate of events is given by the average time to find a proof-of-work hash solution given the current level of mining difficulty and the network’s cumulative hash rate.]
But the Poisson process (i.e. a probabilistic) model is unaware of a partition running (e.g. on Mars by aliens) at a different hashrate. The “independent” random variable in the Poisson model is actually dependent on the probability distribution of partition events (i.e. an orthogonal probabilistic model)— something you will never see mentioned in any mathematical discussion of Bitcoin. Heck this isn’t merely theoretical nor so far-fetched given the radio signal propagation delay between Mars and Earth ranges from 4.3 to 21 minutes, making Bitcoin entirely unsuitable for an interstellar mined blockchain.
[7] In an asynchronous network,[8] footnote #1 in Andrew Poelstra’s On Stake and Consensus (https://download.wpsoftware.net/bitcoin/pos.pdf) correctly cited that absolute termination (i.e. final consistency) is impossible unless **dysfunctionally partition intolerant** (http://web.archive.org/web/20130312052529/http://blog.cloudera.com/blog/2010/04/cap-confusion-problems-with-partition-tolerance/). Nakamoto consensus sidesteps that FLP85 impossibility theorem by (its transactions) being only probabilistically (and thus never absolutely) final.
Proof-of-stake requires a synchronous network assumption enabling “dysfunctionally partition intolerant” deterministic finality (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56466346) but forsaking on-chain objectivity (https://blog.ethereum.org/2014/11/25/proof-stake-learned-love-weak-subjectivity) (c.f. also (https://www.reddit.com/r/CryptoTechnology/comments/od9ves/proof_of_stake_how_i_learned_to_love_weak/) and §4.3 “Long-Range” vs “Short-Range” Attacks (https://download.wpsoftware.net/bitcoin/pos.pdf#subsection.4.3)) inter alia.[9]
[8] Asynchronous networks have no max message arrival latency. Actually Bitcoin’s 10 min block period is a synchrony assumption, but the (propagation delay) network diameter is so much smaller; it’s for practical purposes almost (https://tik-old.ee.ethz.ch/file//49318d3f56c1d525aabf7fda78b23fc0/P2P2013_041.pdf#page=8) equivalent to an asynchronous assumption. C.f. 1 (https://www.reddit.com/r/btc/comments/gxng23/comment/ft97xgd/), 2 (https://www.reddit.com/r/btc/comments/gxng23/comment/ft3qgsi/) and Vitalik’s summary w\less exact equation (https://blog.ethereum.org/2014/07/11/toward-a-12-second-block-time#stales-efficiency-and-centralization).
[9] Specific (https://web.stanford.edu/class/ee374/lec_notes/lec15.pdf#subsection.1.4) (dubiously (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7liveness-attack)-ameliorated (https://web.stanford.edu/class/archive/ee/ee374/ee374.1206/downloads/l18_notes.pdf)) attacks against (https://bitfury.com/content/downloads/pos-vs-pow-1.0.2.pdf#page=16) proof-of-stake consensus distract focus from insoluble deficiencies of lacks cost-of-production (https://steemit.com/bitcoin/@anonymint/secrets-of-bitcoin-s-dystopian-valuation-model) thus no monetary value (https://archive.ph/https://www.zerohedge.com/news/2019-05-25/drop-gold-myths-naturalist-exposition-golds-manifest-superiority-bitcoin-money), inferior liveness (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7liveness-attack) (rubber-hose (https://www.google.com/search?q=rubber-hose+cryptography) the whales![10]), can’t distribute decentralized (https://bitfury.com/content/downloads/pos-vs-pow-1.0.2.pdf#subsection.3.2) and cost-free forks (https://bitfury.com/content/downloads/pos-vs-pow-1.0.2.pdf#subsection.3.1) (lack opportunity cost bcz) are equally (in)“secure.”
I wrote (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7what-might-work):
“stake is always conserved”
[10] https://www.reddit.com/r/programming/comments/7ph6m/rubberhose_cryptanalysis_russian_name/ (Russian name: "Thermorectal cryptanalysis" 🤣)
===========
Hypothetically even brute-force, nearly cost-free, long-range attacks (analogous to a withheld/uncontactable partition) might someday be viable on Bitcoin employing hypothetically[2] a quantum computer employing Grover’s algorithm for preimage search[3] might be able to orphan the entire Bitcoin history since inception, c.f. *§5 Resistance to quantum computations* of Iota’s Tangle whitepaper[4] although quantum computers are no where near practical yet[3][5] and Daniel J. Bernstein showed that some (at least one) quantum algorithms (e.g. hash collisions) will **never** be more cost-efficient with a quantum computer.[3][6]
Note this posited attack presumes the profit/motivation from/for orphaning the non-quantum history subsumes appending quantum computing mined blocks to said history.
[2] https://www.scottaaronson.com/democritus/lec14.html
[3] https://crypto.stackexchange.com/questions/63236/what-benefits-quantum-offer-over-classical-parallelism
[4] https://assets.ctfassets.net/r1dr6vzfxhev/2t4uxvsIqk0EUau6g2sw0g/45eae33637ca92f85dd9f4a3a218e1ec/iota1_4_3.pdf#page=26
[5] https://viterbischool.usc.edu/news/2023/06/quantum-computers-are-better-at-guessing-new-study-demonstrates/
[6] https://cr.yp.to/papers.html#collisioncost (https://cr.yp.to/hash/collisioncost-20090823.pdf)
Reacting to:
https://0xfoobar.substack.com/p/ethereum-proof-of-stake#%C2%A7ethereum-implementation
“Ethereum’s PoS implementation has been teased for the better half of a decade now, but with the beacon chain running for 18 months straight and successful live merges on multiple testnets the initial implementation is largely finalized.”
Dubious whether bugs and corner cases have been exorcised given the high likelihood of the natural oligarchy enforcing a preferred order:
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56417630
“A supermajority (2/3rds) of validators is required to finalize a block, in case of a 50-50 network partition blocks would stop being finalized and attestation rewards would stop. Non-participating validators would slowly leak stake through the inactivity leak until online validators once again had a supermajority. This is the ‘self-healing’ mechanism that allows both safety and liveness.”
THIS ENTIRELY LACKS THE PARTITION TOLERANCE OF THE CAP THEOREM.[1] This design bleeds out the partitioned stake from every partition, so they are permanently forked off from each other unlike for proof-of-work where all but one of the partitions will eventually be orphaned when the network returns to normal.
[1] https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/57176124
Also this presents either an inferior liveness situation to proof-of-work or means for the natural oligarchy (probably insidiously, surreptitiously) extract value:
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56417630
“Equivocation is punished by slashing up to the validator’s entire stake, so the attacker must commit to destroying at least one-third of all staked ETH. The cost to reorg a finalized block is several billion dollars, even at today’s depressed prices.”
Not that they have any incentive to do so when the natural oligarchy (w\majority stake) can more easily and effectively extract value insidiously instead. There are theoretical long-range attacks for said natural oligarchy in proof-of-stake against lock-up deposit penalty mechanisms, but this simply isn’t the preferred way that proof-of-stake systems mine the FOMO n00bs instead by surreptitious value extraction.
“One key difference is that the honest validators would have to explicitly band together to recognize one another’s attestations and override the fork choice rule, but other than that they can form their own child chain and the malicious supermajority would slowly bleed stake out of the validator set until the honest subminority has once again regained a supermajority.”
The majority stake chain can censor blocks and attestations! The bleeding based on inactivity of attestations is a double-edged sword that can be (even insidiously) turned against the subminority by the malevolent supermajority. The “he said, she said” dilemma of weak subjectivity[2] applies if instead the honest minority band together to create their own fork to bleed out what they claim is the dishonest majority.
[2] https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56420675
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56417630
In addition to some additional points I made at the aforelinked, Lyn Alden also refuted (https://www.youtube.com/watch?v=1m12zgJ42dI) the following linked section. Essentially she argued against the highest economies-of-scale have the most proof-of-work profit, because she alleges there’s too much risk (e.g. jurisdictional) and micromanaged opportunity-cost (e.g. making deals in situations where byproduct energy or natural sources are normally discarded/unharvested).
https://0xfoobar.substack.com/p/ethereum-proof-of-stake#%C2%A7pos-rich-get-richer-pow-egalitarian
I reacted to:
“The first time a node comes online, it must have some frame of reference to know how to bootstrap itself. This is not a problem unique to PoS; even a bitcoin full node needs to know which client software is valid, which IPs to bootstrap its history from, etc. PoS adds only minor additional trust assumptions here.”
https://0xfoobar.substack.com/p/ethereum-proof-of-stake/comment/56420421
Disingenuous hyperbole at least w.r.t. practical usability if not also the fundamental invariant.
Trusting your software client to be running the correct protocol is not a subtle distinction from the-human-user instructing the software client which of the SUBJECTIVE conflicting forks is the canonical one. For one because most users will be incapable of doing that without being vulnerable to deception (e.g. phishing attacks, etc).
Fundamentally as you know, there’s no way to ever OBJECTIVELY know for sure what was/is with proof-of-stake (or any non-proof-of-work) consensus system. That is so due to the possibility of a network outages/partitions, even if the user thought their software client was active in real-time for the entire history of the blockchain.
Either of the aforementioned mishaps with proof-of-stake only need to happen once for great harm; whereas, there’s no choosing of IPs with Nakamoto consensus, as the user’s software client can AUTOMATICALLY compare the results from innumerable IPs that claim to connect to the canonical gossip network, OBJECTIVELY choosing the longest (or more accurately the highest cumulative mining difficulty) chain. The exhaustive search for objectively can continue indefinitely, thus highly improbable the software client is going to be remain walled off from the objective reality.
I reacted to:
“That's dangerous to use social consensus to ‘slowly bleed stake out of the [censoring] validator set.’ What if the censoring validator set is doing so because of government mandates…”
https://0xfoobar.substack.com/p/ethereum-proof-of-stake/comment/56416910
Indeed. Or for example if the natural oligarchy censors (i.e. ignores) the minority attestations (making them appear to have never been sent bcz not recorded) thus causing those honest validators to bleed out (due to recorded inactivity) even though they’re not malevolent and were sending attestations across the network. Thus further centralizing the stake (i.e. wealth) of the blockchain in the natural oligarchy. An example of why I had coined/adopted the term ‘Consortium’ blockchains for proof-of-stake:
https://steemit.com/blockchain/@anonymint/consortium-blockchains-e-g-dpos-and-tendermint-can-t-internet-scale
Ethereum’s proof-of-stake (PoS) security theater (e.g. slashing, etc) presumes we’re n00bs unaware of the existence of the natural oligarchy coupled with the fact that “reality” in networks is a blame game (which is precisely what consensus is for, duh), and the majority decides who to blame and thus can paint any reality they want if they can agree to (which is trivial to lie about given the natural oligarchy is the majority). C.f. my references to ‘blame’:
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7natural-oligarchy
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7what-wont-work
Any technobabble minutiae counter argument is refuted w\the inviolable fundamental generative essence fact that the natural oligarchy (i.e. w\majority stake) has total control and change the reality as recorded by the protocol in all “proof-of-stake” (i.e. any not proof-of-work (PoW)) consensus systems (even those w\obfuscating names/concepts that all actually distill to natural oligarchy control, e.g. proof-of-activity, proof-of-importance, proof-of-elapsed-time, proof-of-contribution, proof-of-coverage, etc). Their designers/protagonists are unable/unmotivated to pinpoint a specific mechanism of the vulnerability, it always exists due to said insoluble invariant in non-PoW distributed consensus.
The argument that PoW is also vulnerable to a 50+% (actually less than 50% due to orphan rate) takeover is not in every facet an equivalent vulnerability because expended work, ASIC depreciation and sunk cost are not conserved unlike stake which is always conserved (for the natural oligarchy which can obviate all PoS protocol penalties in their favor).
For example, unlike in PoS, the natural PoW mining oligarchy can’t SUBJECTIVELY vacate liveness w/o forsaking revenue, ASIC depreciation and/or sunk cost. Alternatively can mine empty blocks and (via 50+% attack) orphan all non-empty blocks but this attack is OBJECTIVELY distinguishable from a flaky/partitioned network. Honest miners could choose to objectively fork away by having an automated Schelling point protocol rule to ignore empty blocks. Unfortunately Bitcoin’s PoW is flawed in that the miner’s block could be filled with transactions (txns) to self, paying all the transaction fees to self, which is surreptitiously as deleterious as an empty block. I posited an idea for possible improvement:
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7what-might-work
Note a 50+% attack orphaning all minority block (regardless if empty or not) does unfortunately progress subjectively to 100% centralization of mining, not objectively distinguishable from a flaky/partitioned network (ditto as aforementioned for PoS). This is the selfish-mining attack in extremis in that 100% of the rest of the network’s blocks are orphaned.
However unlike for PoS, there exists (except for CPU-only mining such as Monero’s RandomX) the nuclear *THREAT* (thus likely never necessary!) of hard-forking to change the PoW algorithm to burn all of the attacking mining oligarchy’s ASIC sunk cost instantly turning those ASICs into door stops. What is the equivalent threat in PoS? Burning the oligarchy’s stake will not work. In PoW, the natural mining oligarchy isn’t necessarily the natural token oligarchy. Whereas, in PoS the two are conflated. I wrote:
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7natural-oligarchy (click for links in quoted)
“Intrepid Vitalik’s naivete belies the fact that the ~80 - 99% are the power vacuum—and ‘nature abhors a vacuum and for vacuums of power it’s doubly true’—which are naturally harvested by the ~1 - 20% per the inviolable natural order of the power-law distribution of wealth and fungible resources. Any blockchain which burns ~80 - 99% of its stake (i.e. chops off its face, disembowels itself) to allow its ~1 - 20% economic minority to transact (i.e. to save its skin — the ~80 - 99% majority of thus nearly worthless human protoplasm) will isolate itself from anything economic¹⁵ (i.e. be worthless).
Thus not only would the oligarchy retain economic control on their fork, if there was some inane ‘social consensus’ witch hunt attempt to blame the merely suspicious (but not objectively provable—only subjectively—culpable) validators.”
Said nuclear threat isn’t entirely vacuous bcz the hard fork although potentially initially of much lower security (e.g. if no ASICs yet exist for it, though the economic pressure to create them pronto could be immense), is an air-drop wherein everyone can vote by which chain they sell to buy the other w\proceeds.
Note the block-censoring, PoW natural-mining-oligarchy would decrease the systemic hashrate by that of the censored minority (i.e. up to ~50%) thus presumably decreasing the price but potentially recoverable by the reduced competition and mining difficulty. Yet it’s unlikely this would not be perceived as an attack, thus cratering the price. Ditto for PoS, except that the damage (e.g. as compared to hashrate-driven valuation and security in PoS) reversible (e.g. potentially after profiting on shorting) without any forsaken ASIC depreciation and minority miners destruction (i.e. which do not return). More likely in both PoW and PoS is insidious trickle “selfish-mining” (as a catch all term for block censoring) to gain market share over the long-term and other forms of insidious value extraction[1] such as MEV (with objective randomness being a complete farce on PoS blockchains as I explained in my blog).
Jurisdictional arbitrage likely obviates governments successfully regulating PoS validators or PoW miners unless in some future NWO world government w\the Antichrist as the natural oligarchy. Meanwhile, ownership can be severed from control with appropriate exigent contingency plans for when an owner is arrested, e.g. distributed multi-sig and geographically distributed ASICs.
I’m more worried about insidious censoring of txns aligned to the interests of the natural oligarchy. My aforelinked posited improvement to Bitcoin (and any applicable PoS designs) would make including any txns (i.e. non-empty blocks) an all-of-nothing affair and/or make implausible for miners/validators/block-builders to know the txns before adding them to a “finalized” block. E.g., a superfluous txn and a required supposedly random number in each txn might defeat block builders’ a priori demand for txn deets bcz said number could be the hash of a txn to be revealed later— tho builders might block based on superfluous correlations.
In summary but c.f. my follow-up (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56467747), excluding a separate issue that PoS generates no production cost monetary value and can’t distribute tokens in a decentralized meritorious manner, and assuming the existence of the natural oligarchy, there’s little practical difference in effect between PoW and PoS on the operation and outcomes (although PoS has a severe liveness inferiority if there’s no a natural oligarchy), excepting that the PoS natural oligarchy has more cost-free control (i.e. rope-to-hang-itself-with) thus could f-up, (perhaps irrelevant if the natural oligarchy is preventing all conflicting forks) PoS’s weak subjectivity and lack of non-manipulated public randomness on PoS (i.e. a surreptitious insidious extraction of value[1]):
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56420675
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56415861
Both PoW and PoS will probably become 100% centralized for the aforementioned reasons, although Lyn Alden argues PoW is more difficult to monopolize (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56466346). [1]
https://0xfoobar.substack.com/p/ethereum-proof-of-stake#%C2%A7proof-of-any-resource-is-centralizing
This at least in the near-term probably doesn’t manifest as any incentives incompatible outcome that destroys the value of respective blockchains. When the natural token oligarchy’s interests align with some global enslavement paradigm (e.g. the biblical Antichrist), blockchains will be lights out for human civilization.
https://bitcointalk.org/index.php?topic=160612.0
(Bitcoin: The Digital Kill Switch)
My only hope for averting that eventual dystopian end game is for fungible money itself to become meaningless:
https://bitcointalk.org/index.php?topic=355212.0
http://web.archive.org/web/20121123024723/www.coolpage.com/commentary/economic/shelby/Demise%20of%20Finance,%20Rise%20of%20Knowledge.html
(Demise of Finance, Rise of Knowledge)
[1] Jason Drake admits PoS stakers have huge opportunity cost (https://youtu.be/1m12zgJ42dI?t=5340), thus logically they must surreptitiously extract more value! Largest PoW miners hedging risk reduces their rate of market share accumulation w.r.t. smaller PoW miners who take more risk!
Heck the proof-of-stake opportunity cost includes that staking is just a ruse to trap fools inside the Titanic:
https://cobie.substack.com/p/apecoin-and-the-death-of-staking
Lyn Alden *might* be correct that proof-of-work mining is not 50+% centralized due to the aforementioned current risks (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56417630). But unfortunately under an Antichrist NWO world government, those risks might eventually be vacated by the said supreme power, thus enabling mining centralization. The said nuclear threat might become impotent because the world would be too dependent on Bitcoin by then. Which is exactly the Biblical end game and what I was alluding to with my 2013 Bitcoin: The Digital Kill Switch.
Meanwhile, as Jeff Booth eloquently explains to Green Capital (https://youtu.be/xf4RqFuru2g?t=2913), Bitcoin is the only reliable, non-manipulable, “neutral” (https://youtu.be/xf4RqFuru2g?t=528) unit-of-account “measurement” (https://youtu.be/xf4RqFuru2g?t=761) of complete reset of the global order to one of massive abundance which invalidating (https://youtu.be/xf4RqFuru2g?t=643) the economic viability of the extant political systems (https://youtu.be/xf4RqFuru2g?t=186). Jeff cites (https://youtu.be/xf4RqFuru2g?t=1184) the blockchain trilemma (security, decentralization, scalability) as the reason proof-of-stake can’t have the aforementioned qualities. He first implies that Ethereum is insecure (which it has been but that was due to Turing-complete programmability: http://trilema.com/2016/to-the-dao-and-the-ethereum-community-fuck-you/). I’d argue proof-of-stake is “insecure” because it’s centralized (which Jeff also thinks), e.g. how I explained in this blog that the natural oligarchy can manipulate randomness to manipulate the outcome of the programmable smart contracts. Jeff states, “who would build a decentralized finance on top of a centralized, expensive database, it’s so ludicrous…they don’t have utility…I can’t see one thing that stays valuable in the crypto ecosystem [other than Bitcoin]…create my own coin and pretend it has value…they lobby governments with the money they’re stealing from people…they’re all going to go to zero anyways.” I also predicted that where I wrote in early 2019 (https://steemit.com/blockchain-scaling/@anonymint/lightning-networks-must-fail-if-it-succeeds):
“…(aka block size limit) has to be immutable otherwise the ‘social consensus’ involved to change it would (as democracy and voting always does) steal, squander, and destroy all the value…envision scenarios such as by politicians representing competing forks offering some benefits (http://esr.ibiblio.org/?p=984) …”
Jeff seems to think Lightning Networks (which I explain is ultimately fractional reserves “Mt. Box”) is not centralized crap (https://youtu.be/xf4RqFuru2g?t=1443, https://youtu.be/xf4RqFuru2g?t=2562) thus not a threat to Bitcoin’s security but I explained otherwise (https://steemit.com/blockchain-scaling/@anonymint/lightning-networks-must-fail-if-it-succeeds) including that at least **unbounded transaction scaling is incongruent with fixed token supply in proof-of-work**. And that (among reasons) is why that 2017 soft-fork (which enables the Hash Time Locked Contracts (https://en.bitcoin.it/wiki/Hashed_Timelock_Contracts) required by Lightning Networks) will be destroyed and removed from legacy protocol Bitcoin (https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7what-wont-work). Thus **Jeff is ostensibly unable to see that he will be kicked off of Bitcoin and that he will also be enslaved in the dishonest ledger CBDC paradigm.**
Jeff is failing to incorporate into his futurist, utopian model the inviolable invariants that the power-law distribution dominates fungible resources and that power-vacuums can only be sustainably captured by the most ruthless.
Assuming the existence of a natural oligarchy (which is mostly certainly the case, at least and if only for proof-of-stake), thus overly dramatized/emphasized (aka security theater) concerns about security (c.f. Power Ranger Jason Drake’s pointless and witless nonsense where he conflates security with non-out-of-thin-air monetary value creation: https://youtu.be/1m12zgJ42dI?t=5619) are vacuous/pointless given the natural oligarchy has no incentive to destroy the value of its own blockchain and instead an incentive to insidiously/surreptitiously extract value from it, which is much easier in proof-of-stake. Proof-of-stake exists to fool n00bs and mine the FOMO for value extraction. Because in neither case for Bitcoin or Ethereum, is either at any risk of being 50+% attacked until perhaps FOMO is dead and proof-of-stake has thus reached its utility expiration date. Upstart, fledgling smaller proof-of-work blockchains are more vulnerable to proof-of-work difficulty non-readjustment bomb attacks than proof-of-stake (http://trilema.com/2014/the-woes-of-altcoin-or-why-there-is-no-such-thing-as-cryptocurrencies/), but that’s irrelevant to the salient consideration herein.
Jason’s arguments about the value of the social layer are also vacuous:
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery/comment/56417020
================================================
“…excluding a separate issue that PoS generates no production cost monetary value…”
I wrote elsewhere:
“…lacks cost-of-production (https://steemit.com/bitcoin/@anonymint/secrets-of-bitcoin-s-dystopian-valuation-model) thus no monetary value (https://archive.ph/https://www.zerohedge.com/news/2019-05-25/drop-gold-myths-naturalist-exposition-golds-manifest-superiority-bitcoin-money) …”
Hilarious that Jason Drake argues (https://youtu.be/1m12zgJ42dI?t=4488) that reducing the liquidity of tokens originally created out-of-thin-air[2] in proof-of-stake generates more token value than expending them on external resources in proof-of-work to set the production cost valuation of digital commodity money. It’s no less wasteful than proof-of-work, while not imparting any cost-of-production valuation to digital commodity money:
https://www.truthcoin.info/blog/pow-cheapest/#claims-1-3-4
[2] because PoS can’t with decentralized protocol distribute new tokens in exchange for useful work
Someone commented:
https://www.youtube.com/watch?v=1m12zgJ42dI&lc=UgwV4eqC-vz43GV1-rF4AaABAg
“Something I've noticed with many developers and engineers (as I used to be an engineer) is that they are unable to see the bigger picture. They become so lost in the specific technicalities of a product that their view on macroeconomics is often extremely poor. This is common among most Ethereum advocates. They're all technically very savvy but their knowledge on economics is poor.”
I reacted to:
“Yes, social consensus underlies everything, even choices like which Bitcoin code is the real code”
https://0xfoobar.substack.com/p/ethereum-proof-of-stake/comment/56416242
Nonsense. USAF in 2017 was an attack on the Nash equilibrium thus will not sustain as Nakamoto consensus is only stable in said equilibrium. Specifically the impostor soft-fork created a (by now) ~17+ million BTC booty which means miners no longer have a single best strategy as required by Nash equilibrium. The novel pay-to-script-hash protocol is ANYONECANSPEND (in Satoshi’s legacy protocol) and will be completely fleeced (i.e. taken from wallets w/o touching the private keys because ANYONECANSPEND means a signature w\private key isn’t required to spend it) bankrupting for example BlackRock and all its corporate Wallstreet (even now the Wisconsin government and of course Michael Saylor who I tried to warn via his business mgr but was ignored) minions and sycophants (perhaps at the halving in 2028). The game theory and economics of the situation are inarguable, if you actually are not clueless a.f..
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7what-wont-work
Contrary to Justin’s snowflakes delusion (https://youtu.be/1m12zgJ42dI?t=975), the social layer of proof-of-stake doesn’t override the natural oligarchy in any economic sense. I wrote:
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7natural-oligarchy
“Intrepid Vitalik’s naivete belies the fact that the ~80 - 99% are the power vacuum—and ‘nature abhors a vacuum and for vacuums of power it’s doubly true’—which are naturally harvested by the ~1 - 20% per the inviolable natural order of the power-law distribution of wealth and fungible resources. Any blockchain which burns ~80 - 99% of its stake (i.e. chops off its face, disembowels itself) to allow its ~1 - 20% economic minority to transact (i.e. to save its skin — the ~80 - 99% majority of thus nearly worthless human protoplasm) will isolate itself from anything economic¹⁵ (i.e. be worthless).
Thus not only would the oligarchy retain economic control on their fork, if there was some inane ‘social consensus’ witch hunt attempt to blame the merely suspicious (but not objectively provable—only subjectively—culpable) validators.”
Even the OP blog author seems to be aware of this economic reality:
https://0xfoobar.substack.com/p/ethereum-proof-of-stake#%C2%A7real-world-asset-tokenization-makes-forking-difficult
I commented:
https://0xfoobar.substack.com/p/ethereum-proof-of-stake/comment/56415832
You’re ostensibly writing about topics outside your capability to fully grok the implications of what is actually going on. You should be more circumspect.
ʀᴀɴᴅᴀᴏ can be entirely hijacked by the natural oligarchy (scroll down to the discussion of proof-of-stake):
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7blockchain-randomness
And ʀᴀɴᴅᴀᴏ “appears to be a hack, an ethos of informality preferred by Vitalik”:
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7liveness-attack
ʀᴀɴᴅᴀᴏ has no complete formalization and is heuristic of unknown vulnerabilities, unlike for example Dfinity’s (or Ourboros’) beacons which are formalized (but still suck in terms of liveness and fail to the natural oligarchy).
Proof-of-stake is dogshit and I will make this argument more comprehensively in other follow-up comments on your very myopic blog.
Note you did correctly acknowledge the inviolable existence of a natural oligarchy:
https://0xfoobar.substack.com/p/ethereum-proof-of-stake#%C2%A7proof-of-any-resource-is-centralizing
And I do have a posited solution in mind even if it’s not fully articulated below (yet):
https://anonymint1.substack.com/p/decentralized-data-storage-and-delivery#%C2%A7what-might-work