S17 E38: Keagan McClelland (Keags) on Lightning Metrics, Covenants, Consensus Politics & Verifying Bitcoin

Keagan McClellan (Keags) co-founded Start9 Labs, worked on Sapio with Jeremy Rubin at Judica during the CTV saga, spent time at Lightning Labs maintaining LND, and is now building BTC Verified, a formal specification and verification project for Bitcoin's consensus rules written in Lean 4. He is also a self-described member of Bitcoin's "progressive wing" who nonetheless defends the rising bar for consensus changes, a pluralist who spends all his working hours on Bitcoin, and a security thinker who refused to dunk on Coldcard even while sitting across from someone who very much wanted him to.

Time stamps:

00:00 – Intro and sound check
00:27 – Introducing Keagan "Keags" McClelland
01:14 – Bitcoin origin story: darknet markets and a CS degree in 2015
03:42 – Procrastinating at Amazon by reading ICO white papers
05:11 – Leaving Amazon for Salt Lending
06:24 – How Start9 Labs was founded
07:52 – Leaving Start9 for Jeremy Rubin's Judica (Sapio)
09:47 – What CTV (formerly OP_SECURE_THE_BAG) enables
12:35 – Mastercoin's original pitch, Rare Pepes, and play before serious use
15:44 – The Lindy effect and institutional inertia
18:20 – Sponsors + giveaway rules
22:24 – The road to Lightning Labs and LND
23:55 – Lightning as a complexity class change
25:28 – Data locality, privacy, and incentives in payment channels
27:33 – The 133 MB block problem
30:20 – Scaling takes decades: internet lessons
31:24 – Centralization, Moore's law, and what scaled the web
34:58 – Zcash's Near Intents as outsourced Bitcoin privacy
37:05 – Pluralism and the freedom to transact
39:05 – The gravitational well of money
41:36 – When does a shitcoin stop being a shitcoin?
45:43 – Every soft fork since Satoshi
47:36 – The consensus wall: CTV and BIP300
51:56 – BIP 110 case study: incoherent goals, broken design
56:05 – Why information theory defeats spam filters
58:56 – Node mythology and economic weight
1:05:17 – Miners can activate invisible soft forks
1:07:18 – Luke's soft fork definition and herd immunity
1:10:45 – Chat drama and "how much Bitcoin do you own?"
1:14:15 – RBF, Bitrefill, and policy vs consensus
1:17:21 – Can fees alone secure Bitcoin? The 50% hash rate threshold
1:24:56 – Giveaway reveal: the number is 37
1:26:55 – The Coldcard exploit
1:34:10 – The AI exploit era: "cannot be affected" vs "have not been"
1:39:13 – Why the conspiracy theories teach the wrong lesson
1:42:51 – Against security through obscurity
1:47:27 – Onboarding friction and the paranoia continuum
1:55:42 – Vaults explained, and building them without CTV
2:00:42 – Pre-signed transactions, toxic waste, and Liana
2:05:45 – Would covenants have saved Coldcard users? (No)
2:16:53 – Magic wand: why Simplicity beats CTV and GSR
2:20:33 – Witness size, block size, and what BCH got right
2:24:50 – Why Bitcoin should copy homework
2:29:49 – The two hard forks: BIP 110 vs Ecash
2:33:16 – BTC Verified: formal verification of Bitcoin consensus
2:37:10 – TABConf plans
2:39:23 – Outro and sponsor thanks