Thread

image So why did Luke Dashjr say "Hashrate is irrelevant at this point"? I won't pretend to know what his opinion is, I'll just give you mine. 1) Mining isn't decentralized — pools are (allegedly, not really). Why are pools not decentralized? Because the interests of "rival" countries converge more often than most realize (More context: https://controlplanecapital.com/p/rivalry-between-countries-is-curated ). Hashrate aggregates into ~5–7 top pools; 2–3 could cross 51% at times. Miners "vote with feet", but: - Payout variance pushes them to big pools. - Pools integrate with regulated fiat ramps and insurers; they may adopt OFAC/blacklist templates to avoid headaches. Template power: Pools select transactions. If major pools adopt "policy clients" (e.g., template filters, blacklists), settlement becomes steerable — even if blocks remain valid under consensus. So what? Incentives favor pool compliance with state policy — especially when insurance, utilities, and public listings are involved. 2) Hardware & energy are choke-points (supply-chain centralization) - ASIC oligopoly. 2–3 manufacturers dominate. Firmware signing, remote management, and replacement cycles create vendor leverage. - Jurisdictional energy. Large industrial miners rely on permits, grid interconnects, subsidies. In a low Gross Consent Product environment, regulators swap "ideals" for "stability" — conditional access > rights. - Policy carrot/stick: cheap power for curtailment agreements, transaction policies, ESG attestations; penalties for non-compliant operators. So what? If you need the grid and the power plant, the power plant owns you, and guess who owns the power plant. I have just described the current state of mining "decentralization". Of course this could improve/worsen in the future. Unless mining decentralization improves significantly, increasing hashrate is irrelevant at this point.

Replies (6)

To be fair, I think he’s referring to the issue with Core 30. The mining pool centralization problem is already solved with Datum - it just needs much wider adoption. Core 30 is an entirely different issue. BIP-444 could be a potential solution, but it remains to be seen whether a USAF will gain enough traction and support to make it happen. If it does, I’ll be more bullish than ever.
↩ replying to Tauri
Could be, it is not very relevant what he is referring to, I just analyzed the message. Core 30 opens up major attack vectors as well. But to say "the mining pool centralization problem is already solved with Datum - it just needs much wider adoption", is like saying "the MoE problem is already solved with Bitcoin/Monero - you just have to get people to use them". Inside OCEAN, DATUM is now the default; network-wide it's still ~1% or less. OCEAN is only about ~1% of total Bitcoin hashrate (order of magnitude; varies over time). If ~65–94% of OCEAN’s hashrate is on DATUM, that implies ~0.6–0.9% of the network hashrate is mining with DATUM today. Stratum V2 is available at Braiins Pool and OCEAN/DEMAND’s pool, plus scattered tests elsewhere, but network adoption remains ~1% of hashrate or below. So mining pool centralization isn't a solved problem for now. Solvable problem, but not solved.
↩ replying to Tauri
BTW, there's a reason less than 2% run protocols in which miners, not pools, decide which transactions get added to the blockchain. The Controllers kind of have the miners by the balls. Steering can come from: - Energy & permitting (Power contracts, Carbon accounting credits, Zoning / environmental sign-off), - Banking access, Insurance, Capex financing, - ASIC supply choke: export licenses/customs enforcement prioritize shipments to "certified" operators, - Relay & mempool policy, - Safe-harbor statutes (if you run a certified policy client, you're presumed compliant. If you don't, you own the legal risk), - Reporting/KYC on pool payouts: force miners into KYC'd payout rails; non-KYC payouts trigger Suspicious Activity Reports and audits, - Tax policy: extra depreciation/credits for compliant miners; audits for the rest. And these are just a few of the choke-points. So Bitcoin mining is a tough gig.
I think what Luke is referring to as a weakness is the insider attack vector. Developers with misaligned fiat incentives. Hashrate doesn't protect against bad code/policy decisions. It also doesn't protect against the attack vector enabled by said code/policy decisions. It's not a technical attack but a social attack. Bitcoin's immune system in this case is node runners not running the offensive code/policies.
🛡️
Seems more likely that 'Bitcoin is dead unless we do this' being his opinion is behind it. I hope when this blows over he gets a retreat or something. I love what he's done with Knots and Ocean but this BIP looks like a psychotic break.