Nostr Web Client

There's an interesting general principle at work here;

> "Furthermore, it's designed for this particular use-case and hence

cryptocurrency miners are not incentivized to make optimized ASICs for it.

and yet, how long before someone creates a ccy with *exactly* this PoW? It would result in the amusing situation that people could justify building specialized mining machines precisely because this PoW has the "alternative use case" that so many people in the early days use to crave (remember PrimeCoin). And the idea of "GPU resistance" is somewhere between really dodgy and just nonsense, right.

I had mixed feelings when I found out they were heading in this direction. As I've argued before and I know many others came to a similar place, PoW/hashcash doesn't quite fit the required asymmetry between attackers and defenders (if they are users) (whereas in bitcoin it's not the same because, as it evolved, the users were not the defenders); although it (asymmetry) exists, because attackers need more, it's arguable whether that difference cuts it. And PoW goes against the grain of "UX".

But mixed because: yes, at least in principle, it is *something* and it could actually work, somewhat, especially at the beginning.

Peter Todd 2y ago

Tor doesn't have the same tradeoffs as a crypto coin. Their PoW scheme is versioned, so they can easily update it as necessary and roll out new versions as needed. Tor is highly centralized and fairly trusted anyway so updating to new PoWs as necessary is fine. And of course, the fact that they can do this discourages attackers from even trying to build ASICs.

Reply to this note

Please Login to reply.

Discussion

waxwing 2y ago

That's a good point about centralization. For example Tor Browser forces updates (almost ridiculously often!), so indeed since Tor themselves would be implementing the pow code, there's a ton less lock in than (in the extreme) bitcoin. I do slightly wonder about e.g. random smaller apps (for example, on android? not sure) out there using fixed versions of tor (rather than tbb), i.e. they might become unusable for some applications with a pow change. I suppose it comes to the same thing - you *have* to update according to tor devs schedule (the whole auto-update thing in TBB is really dodgy now I think about it ... I guess it's related to it being Firefox ESR or something?).