Oceans of Wisdom

10% of global TDH gets you done in 3 hours, 1% in a week, 10% of last 10 memes "at any point"

or maybe it can be lower if it takes longer

yeah something like that

So the plan for the quarter is the following: July/August: a) get Stream live b) do the tweetstorms as a NextGen to get them on-chain as opposed to "on twitter" Sept: a) specific plan for eliminating any single points of failure and external dependencies across the network. For example: we still use the Manifold front-end for minting on The Memes. We will need to recreate this functionality on 6529.io and core so that it works fully decentralized We won't have all the single points of failures eliminated by September but I want to have them all mapped out and with proposed solutions for each b) ZKP based 6529 network. This is the final boss. This is clearly "the way" and it is the only part of this list that I do not yet have it all clear in my head. All the rest have been clear in my head since last year, but I wanted to see the The Memes (decentralized) get to some level of maturity before moving to the next steps. We are there, The Memes are working well without me, we can take the next steps now. In the meantime, we deal with bugs, code quality, improvements as they emerge but I think basically the front-end works now, so the main focus shifts to the above

So the plan for the quarter is the following: July/August: a) get Stream live b) do the tweetstorms as a NextGen to get them on-chain as opposed to "on twitter" Sept: a) specific plan for eliminating any single points of failure and external dependencies across the network. For example: we still use the Manifold front-end for minting on The Memes. We will need to recreate this functionality on 6529.io and core so that it works fully decentralized We won't have all the single points of failures eliminated by September but I want to have them all mapped out and with proposed solutions for each b) ZKP based 6529 network. This is the final boss. This is clearly "the way" and it is the only part of this list that I do not yet have it all clear in my head. All the rest have been clear in my head since last year, but I wanted to see the The Memes (decentralized) get to some level of maturity before moving to the next steps. We are there, The Memes are working well without me, we can take the next steps now. In the meantime, we deal with bugs, code quality, improvements as they emerge but I think basically the front-end works now, so the main focus shifts to the above

So the plan for the quarter is the following: July/August: a) get Stream live b) do the tweetstorms as a NextGen to get them on-chain as opposed to "on twitter" Sept: a) specific plan for eliminating any single points of failure and external dependencies across the network. For example: we still use the Manifold front-end for minting on The Memes. We will need to recreate this functionality on 6529.io and core so that it works fully decentralized We won't have all the single points of failures eliminated by September but I want to have them all mapped out and with proposed solutions for each b) ZKP based 6529 network. This is the final boss. This is clearly "the way" and it is the only part of this list that I do not yet have it all clear in my head. All the rest have been clear in my head since last year, but I wanted to see the The Memes (decentralized) get to some level of maturity before moving to the next steps. We are there, The Memes are working well without me, we can take the next steps now. In the meantime, we deal with bugs, code quality, improvements as they emerge but I think basically the front-end works now, so the main focus shifts to the above

lololololol

yup

exactly

i dunno something

maybe 0.5%

but clearly must be lower than the memes

since it is decentralized

we need some quality control

something like that the % are a first draft

yes!!!!

yes exactly! there are many things that 1, 2, 5, 10 people might want to buy but not 300!

I think now we have it working well so we can proceed

but I figured this needed to come after we get The Memes working well

the contract is basically done

been sitting on this for a year almost!

https://docs.google.com/document/d/1mcicMd9bv-SbO2PIrA0ng6hHP3CgykbY0g-79bMXeHE/edit?tab=t.0

for your comments

Proposed approach for Stream

[https://docs.google.com/document/d/1mcicMd9bv-SbO2PIrA0ng6hHP3CgykbY0g-79bMXeHE/edit?usp=sharing](https://docs.google.com/document/d/1mcicMd9bv-SbO2PIrA0ng6hHP3CgykbY0g-79bMXeHE/edit?usp=sharing)

Proposed approach for Stream for comments

oops wrong place

for comments

https://docs.google.com/document/d/1mcicMd9bv-SbO2PIrA0ng6hHP3CgykbY0g-79bMXeHE/edit?usp=sharing

Proposed approach for Stream

or maybe the people who voted did buy and the others did not

fascinating

fasciating!

After being rugged by twitter, I buckled down and re-wrote it

https://x.com/punk6529/status/1943053602853187609

i did a whole thread on this!

privacy and scaling

but they solve multiple problems at the same time

bc they are odd and weird

is building our brains in this direction

and it is the biggest challenge I / we currently have

I am completely convinced that ZKP will play a big role in our (6529's) future

proving cost would be low

our blocks would not be very large at least at first

really, really, really interesting idea

@[simo] @[GelatoGenesis]

This is a very interesting idea

Below is a “long-form translation” of Georgios Konstantopoulos’ tweets into plainer (but still technical) language, plus extra context for how this would work inside Ethereum. ## 1. What “SNARK PoW” actually is Classic designGeorgios’ proposal**Proof-of-Stake (PoS)**: validators lock ETH and *vote* on blocks. The fork-choice rule is *LMD-GHOST*; every block carries thousands of attestation messages.**Proof-of-Work over SNARKs**: the *work* is producing a zero-knowledge proof that the entire block’s transactions are valid. Whoever delivers the first valid proof wins the block. The idea is not to roll back to Bitcoin-style nonce-grinding. Instead, miners (call them **provers**) spend their GPU/ASIC cycles on generating zk-SNARK proofs. Those proofs are *needed* anyway to let the network trust gigabyte-sized blocks without re-executing them. This is a flavour of **“proof-of-useful work”** already researched by Aleo’s **Proof-of-Succinct-Work** and similar systems([Daonft](https://daonft.medium.com/aleos-consensus-mechanism-an-in-depth-look-at-proof-of-succinct-work-posw-659746c38c0c?utm_source=chatgpt.com "Aleo's Consensus Mechanism: An In-Depth Look at Proof of ... - daonft"), [Aleo](https://www.aleo.org/post/proof-of-stake-vs-proof-of-work/?utm_source=chatgpt.com "Proof of Stake vs Proof of Work: A guide to consensus mechanisms")). ## 2. Why Georgios claims it beats today’s PoS Tweet bulletWhat it means (with links)**“higher gas limit because no re-execution”**In PoS Ethereum every full node re-executes *all* transactions to be certain the block is valid. If the block instead ships a SNARK, a node only has to verify one constant-size proof (tens of milliseconds). That removes today’s ~30 Mgas ceiling and opens the door to “Gigagas” blocks. ([X](https://x.com/gakonst/status/1942719141992919262?utm_source=chatgpt.com ""SNARK PoW is more efficient than PoS. - higher gas limit bc no re ..."))**“faster blocks – less networking, no voting”**LMD-GHOST relies on every active validator gossipping their latest vote; that traffic is what caps slot time at ~12 s. In PoW there is only the winning header, so blocks can be propagated and accepted much faster. ([eth2book.info](https://eth2book.info/latest/part2/consensus/lmd_ghost/?utm_source=chatgpt.com "2.3.3 LMD Ghost - Upgrading Ethereum - The Eth2 Book"))**“zero amortised wasted energy/work via IVC”**Incrementally-Verifiable Computation (IVC) lets provers *roll* proofs: each block’s proof folds in the previous one. All cycles go to something the chain would need anyway, instead of brute-force hashing. ([veridise.com](https://veridise.com/blog/learn-blockchain/recursive-snarks-and-incrementally-verifiable-computation-ivc/ "Recursive SNARKs and incrementally verifiable computation (IVC) | Smart contract audits from Veridise"))**“thinner, permissionless MEV pipeline”**Because the work winner is whoever proves the whole block first, you don’t need today’s MEV-Boost relays that coordinate builders ↔ proposers. Anyone can build/prove/submit.**“adaptive difficulty / throughput”**Difficulty = “how big a proof must you finish first”. The protocol can raise or lower that in real time to target a block interval even as ZK hardware improves.**“we will get zkEVM ASICs”**Exactly like SHA-256 PoW produced specialised silicon, open competition to prove EVM circuits fastest should lead to purpose-built zkEVM chips, accelerating the entire ZK ecosystem. ([rigsmineria.com](https://www.rigsmineria.com/en/guia-definitiva-aleo-mineros-rentables-2025/?srsltid=AfmBOoogUN43irW-4szm1mIGQi9XKX8PJOijJDFqVcJhWk_FNrhSr6wZ&utm_source=chatgpt.com "The Ultimate Guide to Aleo: The Most Profitable ASIC Miners in 2025")) ## 3. How the consensus layer would look 1. **Fork-choice** *Replace LMD-GHOST*. A block’s weight is the cumulative SNARK-PoW behind it; the heaviest-work chain wins (like Nakamoto consensus). No attestations. 2. **Finality gadget** You can still keep a small PoS committee (Casper-FFG style) to sign checkpoints every N blocks for faster economic finality. The committee could be selected *per block* using randomness derived from the PoW winner, so stake becomes a thin BFT wrapper. ([eth2book.info](https://eth2book.info/latest/part2/consensus/overview/?utm_source=chatgpt.com "2.3.2 Overview - Upgrading Ethereum")) 3. **Rewards & issuance** The protocol pays the prover out of: - newly-minted ETH (like block subsidies today), - priority fees, and - any MEV they can capture inside the block. Georgios’ “out of pocket” tweet underscores that these payouts must be enshrined at protocol level, not handed out via grants. ([X](https://x.com/gakonst/status/1942719141992919262?utm_source=chatgpt.com ""SNARK PoW is more efficient than PoS. - higher gas limit bc no re ...")) ## 4. Is it “going back to PoW” and what about energy? - **Yes** – it re-introduces an external resource (compute) race each block. - **But** the cycles are *useful*: every joule of energy produces a proof the chain needs. Hash-based PoW produces a number you throw away. Comparative studies show that zk-proof generation is far more *joules-per-transaction* efficient than SHA-256 mining, though it is still higher than PoS, where only verification is on-chain([Coincub](https://coincub.com/proof-of-work-proof-of-stake/?utm_source=chatgpt.com "Proof of Work vs. Proof of Stake: Key Differences Explained | Coincub")). The bet is that the scalability gain (100-1000× more gas per block) amortises that cost better than current rollup-plus-merge-mined architectures. ## 5. Open challenges CategoryWhy it’s hard**Proof speed vs block time**Today a full EVM-equivalent proof for 30 Mgas takes minutes on big servers. Hitting <5 s block times needs 50-100× faster provers or aggressive recursion/parallelism.**Verification cost**Even succinct proofs cost ~300k gas to verify. At Gigagas scale that’s small, but the precompile still needs to be part of the hard fork.**DoS & worst-case proving**Malicious builders might craft pathological circuits that stall honest provers. The difficulty adjustment must detect and penalise that quickly.**Economic analysis**Issuance must be high enough to pay provers *and* maintain chain security, but not so high it wrecks ETH economics. No live chain has run this experiment yet.**Hardware centralisation**ASIC-based proving farms could recreate mining-pool dynamics unless there is ample competition and open-hardware efforts. ## 6. Why Ethereum is the natural testbed - **Liquidity & user demand** – plenty of fees/MEV to pay provers. - **Mature tooling** – EVM circuits, Plonky2/Halo2/Nova libraries, and a deep bench of protocol engineers. - **Road-map synergy** – after *EIP-4844* and *Verkle trees*, the next bottleneck is execution-layer throughput. A SNARK-PoW fork-choice directly addresses that without new L2s. ### Bottom line Georgios is advocating a *radical simplification*: > “Turn the *entirety* of block production into one race – first to prove a huge EVM block wins. Let stake handle only checkpoint finality. Everything else (throughput, MEV, hardware R&D) will self-optimise around that single, paid PoW function.” It’s ambitious, unproven, and energy-heavier than PoS-only Ethereum, but if the technical hurdles fall, it could unlock block sizes and confirmation times that today’s rollup-centric roadmap can’t reach. **Further reading** - Veridise blog – *Recursive SNARKs & IVC*([veridise.com](https://veridise.com/blog/learn-blockchain/recursive-snarks-and-incrementally-verifiable-computation-ivc/ "Recursive SNARKs and incrementally verifiable computation (IVC) | Smart contract audits from Veridise")) - Aleo – *Proof-of-Succinct-Work* white-papers and miners’ guides([Daonft](https://daonft.medium.com/aleos-consensus-mechanism-an-in-depth-look-at-proof-of-succinct-work-posw-659746c38c0c?utm_source=chatgpt.com "Aleo's Consensus Mechanism: An In-Depth Look at Proof of ... - daonft"), [rigsmineria.com](https://www.rigsmineria.com/en/guia-definitiva-aleo-mineros-rentables-2025/?srsltid=AfmBOoogUN43irW-4szm1mIGQi9XKX8PJOijJDFqVcJhWk_FNrhSr6wZ&utm_source=chatgpt.com "The Ultimate Guide to Aleo: The Most Profitable ASIC Miners in 2025")) - Eth2Book – detailed explanation of LMD-GHOST and Casper FFG([eth2book.info](https://eth2book.info/latest/part2/consensus/lmd_ghost/?utm_source=chatgpt.com "2.3.3 LMD Ghost - Upgrading Ethereum - The Eth2 Book"), [eth2book.info](https://eth2book.info/latest/part2/consensus/overview/?utm_source=chatgpt.com "2.3.2 Overview - Upgrading Ethereum"))

figuring out the right use for rep vs TDH is a good project

back to the mines for me!