>But there are a literally limitless number of way to extend the trust of the base layer, up to any and all other layers. Ark is a great example, all transactions off chain, every single user has unilateral exit ability.
Unilateral exit will be too expensive for most users to enforce. I don't think this is the framing that should be used.
There is simply no way to scale the trust assumptions of the base layer to every other layers.
Tradeoffs are necessary and desirable, anything else leaves us operating inefficiently and burdens individuals with undue costs.
Do you have to go to court and spend thousands of dollars to get security in your donut purchase?
The court is never for everyday, small interactions. Neither will the base layer be.
To suggest we cannot extend its trust because fees are too expensive, is to say we can’t have a sustainable society even if we have a just and incorruptible court system, because not everyone can afford to dispute every transaction in front of a judge.
It simply means the cost will be relative to how bad the conflict is, which seems natural to me.
And even that conclusion assumes there’s no better tools that we haven’t discovered yet on how to batch/combine “settlements” across separate networks. I don’t think such a wall exists. 🤷🏻♂️
All I’m saying is that extending trustlessness to every interaction is unecessary if not counterproductive but maybe we agree there.
I do agree on the second part though. There isn’t a problem with trade offs, but I think extending trust upward to other layers isn’t about extending it in totality (if that’s how I worded it), but with the specific trade offs that fits each users or institutions situation and environment.
I think those trade offs and trust assumptions will simply change a lot based on the amount of risk (and thus amount of value), which seems also to make obvious sense.
Oddbean new post about login | logout