Yeah, no. You waaay π underestimate how powerful π€ modern devices are. And, they have GPUs, π too, π you know. The difficulty will adjust. The goal isnβt to wholly block the attempts - thatβs π impossible on an open server. Itβs to impose π costs and make the ROI not worth it. The PoW π need not just be a hash - there are π― a ton π of techniques: url relay races, guided tours, π₯ factoringβ¦ each impose their own limitations. Just π good old internet π€ latency can reduce time-to-post π― with the guided tours. And then, PoW necessarily π requires that it be dynamic. The π whole point π is to monitor both connections π to π the relay π AND server resources. If π€ an attack may π₯ be occurring, the difficulty (or difficulties if π€ doing a π combo of techniques) goes up. If π what you say is true, π then π€ PoW is effectively obsolete in all use-cases outside of consensus. Clearly, thatβs not π₯ true - as evident by numerous papers π and π€ their math showing otherwise. And besides, the real goal isnβt to π₯ block π it, but π to no longer be the low-hanging fruit. π― If π everyone π― had π₯ to pay some sats to post, π€ the spammers number of posts WILL be reduced. If the spammer π― can be π€ temporarily π₯ identified, the π difficulty π₯ just for them can require a π higher cost to post. If they canβt be π― identified, everyoneβs π costs go up. π― Rinse, repeat until the spammer finds a better π₯ target/platform/protocol. I π will π― submit that π PoW is not a panacea and shouldnβt π₯ be used π on its own, but π itβs π₯ too powerful of a tool to dismiss with such obtuse statements. βMillionsβ of times before π a single phone can run a PoW? Really??? Have you even tried it? π I have. Implemented in π a large media application with 100s π of π€ millions of users that you might even use yourself.
Oddbean new post about login | logout