Maybe I'll watch someday how he "refuted LN". So far I'm too skeptical of the claim to even prioritize any verification. LN is basically just the acknowledgment of the following facts: 1) global consensus "blockchains" can't scale (and on top of that they are terrible for privacy, instant UX, MEVil/censorship/spam resistance, etc.), 2) relegating a global consensus "blokchain" to a very slow/expensive last-resort settlement layer, moving most of the economy to local consensus pools, is the only acceptable trade-off considering 1, 3) local consensus pools to realize 2 which require less trust (ie: Poon-Dryja channels) are safer but more expensive (in terms of money, time and UX) than local consensus pools to realize 2 with require more trust (ie: Finney-style credit), 4) we don't really have to chose a single answer to 3, since different local consensus pools can use different security/cost trade-offs, including a rich scale of greys between the two extremes (eg: credit->cashu->fedi->liquid->spark->ark->dryja-poon), 5) the rich patchwork of local consensus pools emerging from 4 would be messy and unusable without a single, global, unified language for invoicing and for swapping/hopping across pools: the Lightning Network. Fin. Not much to refute here.