So yeah… having low-quality nodes would be a lot of work (to fake all the metrics) and for some things even isn’t possible - such as for the trustworthy metrics, where performance of a node gets evaluated by its the peers.
I’ve been telling this and I will keep telling this no matter what others say. TEE is not a silver bullet. TEE has been broken multiple times in the past, and will be broken again, especially given sufficient financial incentive. So TEE raises the bar, nothing more and nothing less. It also severely increases the operational complexity. So that has to be carefully evaluated and taken into account.
Been meaning to respond to this for a while. I don’t think that anyone that is wanting access to TEE’s thinks that they are a silver bullet for everything. I understand why security experts, computer science PhD’s, And extremely smart people would be wary of them. The fact of the matter is that, in general, the enterprise world considers them to be"good enough", And in general, using them is considered to To check all kinds of boxes around due diligence and adequate security protocols, according to all kinds of regulatory restrictions across the globe. There’s literally billions of dollars waiting to be deployed onto the tech that figures them out.
In addition, we use TEEs for internet identity all the time. Why are those enclaves OK but others aren’t? Complexity? Probably so. But still, making them available and using them for critical infrastructure are two very different things.
The sooner we get them, the sooner we can start competing outside of Web2 where the infrastructure and tooling out competes what we have for the IC.
This is totally not relevant to the current conversation, but whenever I see this TEE aren’t perfect discussion I think it’s worth bringing up that they are likely good enough for a lot of application.
