RE: Things I think are interesting in crypto right now - How about you?
No problem! My favorite post on the subject is this one. Here is one of the more important excerpts:
The answer is that segwit uses what is, in my professional opinion, an unthinkably nasty hack: according to the Bitcoin protocol, all segwit transactions can be spent by anybody with no proof of authorization. That means that literally anyone can make a transaction that spends the bitcoins in a segwit address. The "witness data" (which is the rules for who's allowed to claim the outputs, and the proof that the rules were followed on the inputs that claim the outputs) is moved out of the main block and replaced with a simple "anyone can spend me, there are no rules and no proofs," and it's up to miners to know what the rules really are, ignore the "anyone can spend me" instruction, and instead apply the rules from the witness data....
The second implication is even more chilling: for the first time, the Bitcoin protocol will have official support for theft of bitcoins. See, the formal rules of bitcoin will be "anyone can spend these coins without authorization," but the segwit rules will say otherwise. This means that anyone who can get a majority of miners to agree to look the other way (i.e. ignore the segwit rules) can send a transaction that spends those bitcoins without authorization, and it will be valid by protocol. At present, if the majority of miners attempted to do this, their blocks would be invalid (and ignored by the network) because they violate the Bitcoin protocol, but with segwit, the Bitcoin protocol will only understand "anyone can spend these coins" so as long as the miners agree to ignore the segwit rules, the transaction will pass checks. Obviously this means that the miners could collude to steal everyone's segwit balances, but more likely, a government could force mining organizations to allow their transactions to steal bitcoins, reverse transactions, or whatever else. Whereas previously this meddling would've been extremely noisy, causing a hard fork (which would have likely been ignored). With segwit, however, theft by miner collusion is a first class feature.
I like this post because a lot of people just complain about the "lightning network" but that isn't really the main issue. The biggest issue is the fact that there was a change at the protocol level that allows cheating if enough miners agree to "look the other way."
I would recommend reading the entire article though. Enjoy! :)
Is it realistic for miners to "look the other way" on a large enough scale for this to happen? When I hear people like "Ivan on Tech" explain Segwit, it just sounds like a reasonable scaling solution. I will have to more thoroughly read the link you shared.
(
I think there will be better options out there anyway but I think the concern is that it is "possible" for them to do it even if it's not likely. Like I said, I think the solution will be something different. Segwit seems too complicated (and frankly unnecessary) for it to be a good solution and just raising the block size (bch) only solves part of bitcoin's problem.