Nostr Web Client

Federated multisig -> known set of custodians (can't be anon because it would make it vulnerable to a sybill attack) easy for the government to coerce. yes, even in multiple jurisdictions since governments today act in lockstep

Drivechain -> Sybill resistant, annonymous, dynamic, and incentive aligned custodians. Very hard for the governmenr to identify.

Obviously drivechains are superior, how do you not see this?

Guy Swann 2y ago

This isn’t even half of the issue though. Thinking that these are the only considerations is foolish:

• federations can be spread out across jurisdictions. Attacks are permissioned

• attacks on a hash lock are permissionless. Anyone can do it and they don’t have to have any stake in the sidechain whatever

• NONE of this does anything to solve the validation problem, because the only way to assume it’s secure, is if miners are validating everything, and if the only reason it is beneficial is because they don’t have to, then it’s not remotely secure.

• it seems that any attacker can issue a fraudulent transaction from the DC hash lock. And miners are expected to vote against it. Except that this again, presupposes they *are* validating every single drivechain.

— in other words we’ve dumped the validation problem of the *entire* DC ecosystem into their lap, said “don’t worry your don’t have to validate,” but then turn right around and demand they be the judge regarding what is and isn’t valid… demanding that they validate.

• a permissionless attacker in DC can attack literally *every* single DC that exists at the exact same time, with no cost to expanding the attack except the rejection of their block *if* other miners are validating the DCs. Except that if they aren’t, they may very well get continuous “yes” votes from major pools that are just default not caring and want to get fees, which immediately creates an enormous consensus battle that takes place entirely our raise of the chain - which is the whole situation we are trying to avoid.

• lastly, if we take the simple incentive that miners will maximize profits - the claim made for why they will force DCs on the network without consensus - then there is no reason for them to not steal from the sidechains. As participation is again permissionless, and it would absolutely maximize output. We are requiring *social* constraints to keep them honest, not validation rules - because theft is valid in the DC system.

This isn’t remotely as simple as proponents are claiming, and the complete refusal to admit significant trade offs is what has made me so annoyed. And like I said before, it’s not worse than a permissioned federation but it’s also mot clear that it’s any better at scale, and there is no solution to the validation problem here, it’s just *all* being dumped on miners and then hand waved away.

Reply to this note

Please Login to reply.

Discussion

j 2y ago

There may be some good points here but it's obvious you didn't read my much shorter reply (I anticipated the common cope of federation members being in different jutisdictions and why it doesn't actually help much) so I wont engage further.

For Guy's audience: the criticism above (that miners have to validate all sidechains to downvote obvious attacks) is equally baseless.

For example, a miner could just run a light client for the drivechain or even just check a drivechain block explorer to see that a "permissionless" attack is just noise. Of course, this could be easily automated. There is no risk of ddos because these "attacks" would have to pay transaction fees.

Even if there is a light client or block explorer issue, and a miner votes incorrectly, the most likely outcome would be that the issue would be fixed by the people running the sidechain and the rest of the vptes wpyld be correct.

I hope you all can see how easy it was to refute Guy's supposedly well-thought out fud and you consider this exchange next time this influencer tells you you can't have a zcash sidechain because reasons.

Guy Swann 2y ago

If you can’t have a conversation without cheap insults and dishonesty, then how could you possibly expect me to take you seriously?

I’ve gone out of my way to give DCs a lengthy and fair shake. But the proponents consistently refuse to discuss trade offs. Say anything about them negative, admit the obvious fact that there ARE trade offs, and suddenly I’m just an INfLuEnCoOr.

I’d be vastly more likely to continue discussion if I had any reason to think the opinion in favor of it were *sober,* but the amount of hand waving and “they can just run light clients” as if this completely negates the numerous issues I brought up is what makes this conversation impossible.

Anyone who bothered to listen to my show and the ridiculous number of episodes I’ve dedicated to going back and forth on this would know the truth. But if a cheap opinion of *me* is the best thing you have for arguing against my points, then there’s no progress to be made here.

I have no reason to continue a conversation with someone who can’t make a point without elementary school name calling.

Thanks anyway, hope you have a good day regardless.

j 2y ago

Unless "influencer" is an insult, I don't think I used any insults or name calling. So, I don't know what you're talking about.

I like your show tho, so keep it up. The Thunder episode was good.

Guy Swann 2y ago

I have a hard time believing that last sentence wasn’t meant to be entirely derogatory - But regardless I’m glad you liked the Thunder Network episode.

Like I’ve said probably 100 times before, DCs are an interesting addition to the architectures of sidechains, but they simply aren’t a panacea, and it bothers me that people act like they are. Makes me feel like I’m not able to get honest, adversarial thinking discussion.

j 2y ago

Fair enough.

Agree that more adversarial discussions is a good thing for everyone.

mleku 2y ago

what exactly is interesting about adding another blockchain onto a blockchain? never heard of Cosmos? you can have whole webs of them if you want. probably you never heard of them and that sorta suggests there's nothing useful really going on.

mleku 2y ago

finality and atomic swaps built in don't fit well with probabalistic finality either, it's already slow af with cosmos tokens crossing between chains, needs at least one hour and even then. someone made finalizer widgets for bitcoin at one point i think.

it's really a dead end, people have been up and down that road at least 3 times already, why now? probably to pump someone's bags.

and as a result of hearing mr guy swann saying affirmative things about old hat says to me there is nothing more worth hearing from him.

j 2y ago

lol, how would you "pump someone's bags" with drivechains? The whole point is to use Bitcoin.

This is just more of the "shitcoins on Bitcoin" nonsense.

As far as Cosmos is concerned, no, I don't know anything about it, I will look into it, thanks. But your point about cross chain swaps not working is silly. Already works today, and there's many account-less exchanges that offer them. The main problem is you'd have to trade your Bitcoin for some shitcoin (except for submarine swaps which are great) so I wouldn't use most of them.

Guy Swann 2y ago

Agreed. The incessant “shitcoins on bitcoin” remarks really misses the whole point. None of what happens on the sidechain matters to the main chain, this is true for literally any of them. All of them could create as many arbitrary tokens if they wanted to.

All that matters is what’s happening on the actual Bitcoin timechain, and how ownership on Bitcoin is determined. Is it a federated multisig, or is it a time locked channel, or is it a hash locked balance? Everything else is completely irrelevant, imo.