> if you take your DA challenge idea to its extreme you will find that it devolves to exactly always embedding blocks on Bitcoin I disagree for reasons about to be discussed > either there is no cost for the challenge in which case I will personally challenge everything just to prove a point That seems unworkable and I think we agree on that > OR there is a high cost for challenging, which can't be refunded because you can never prove the data was withheld so you will just have to pay it That seems reasonable and does not result in always embedding blocks on bitcoin. Users select a sequencer optimistically. If he attempts to withhold data, users can pay a higher-than-normal cost, which is effectively the cost of a unilateral exit. The unilateral exit sequence forces the sequencer to reveal the data he tried to withhold or get slashed. Either way, the users exit unilaterally. That bankrupts the bad sequencer, because all his users unilaterally exited to avoid being victims of the data withholding attack. Then those users pick a new sequencer. Eventually, this process leads to only "good" sequencers remaining. Due to incentive alignment, the process will probably "begin" with only good sequencers as well, since no one wants to go bankrupt, and the only way *not* to be bankrupted is to start out as a good sequencer and remain good. And the great thing about good sequencers is, they don't need to be challenged, because they don't withhold valid data, so no extra "challenge" data goes on-chain -- or, if it does, it's because of a troll user, who pays the price for trolling due to the higher-than-normal cost of initiating the unilateral exit sequence. Thus only trolls pay that cost, which is desirable, as that eventually bankrupts them, leaving only "good" users and "good" sequencers. > So if you want a chain to do things that require people to be aware of what is going on...then you will need to keep posting the data all the time I think the above analysis demonstrates why I think that is incorrect. You can align the incentives properly to bankrupt "bad" sequencers and trolls, such that only "good" sequencers and "good" users are left -- i.e. the ones who follow the protocol in part to avoid going bankrupt. > No Stablecoins backed by Bitcoin are not mere tokens... There is a whole lending contract necessary to mint them, not just a token you mint and transact with RGB without need for observable chain. I didn't say bitcoin-backed stablecoins are "mere" tokens. I just said they are tokens, which you seem to admit.