5 ms·
It turns out that "distributed linked list" is actually a difficult problem that involves very interesting challenges
by bowsamic 4y ago
It turns out that "distributed linked list" is actually a difficult problem that involves very interesting challenges
- LtWorf 4y agoThe "innovation" in the original blockchain is that it is computationally expensive on purpose, to create "economic value". There is no computer science innovation there. Computer scientists didn't come up with the idea because it made no sense. And it all went downhill from there.
- bowsamic 4y agoNo, the goal wasn't to create economic value. The goal was to make it prohibitively expensive to recreate the chain and thus fool someone else. Satoshi did not say that the purpose of PoW was to "create economic value".
- LtWorf 4y agoAnd a digital signature couldn't be used because?
- bowsamic 4y agoA digital signature alone cannot prevent double spending
- LtWorf 4y agoNo, going back in the list of transactions can.
- bowsamic 4y ago
- LtWorf 4y agoProof of work blockchain is nothing technically complicated. Nobody was doing it before because it makes no sense technically. The only reason it is used is because it adds economic value. Anyone using it for other purposes could use something else but wants VC money.
- bowsamic 4y agoOkay then tell me how you can make a distributed ledger with global consensus without a consensus mechanism like PoW or PoS even with “no economic value”
- LtWorf 4y agoConsensus doesn't need to involve doing hashes… there's a million distributed algorithms to do this stuff. it's well researched https://en.wikipedia.org/wiki/Byzantine_fault https://en.wikipedia.org/wiki/Byzantine_fault
- bowsamic 4y agoBut those other solutions aren't just "using signatures". Again, _please point out how you can have a distributed ledger with global consensus with only signatures and no other technology_
- 0xrips 4y agonot only is this provably wrong, but the entire point is also negated through the merge which this post is about.
- morelisp 4y agoBut cryptocurrency doesn't really solve these in a technically interesting way, as it's neither consistent nor available under partition. The pressure to keep the chain consistent and unified is a purely social one - your BTC is only valuable to other people on the same chain as you.
- bowsamic 4y agoI don't know, looking into the papers that are written in crypto research, especially in academia, it seems like there is a lot of very technically interesting stuff going on, especially with zero knowledge proofs for example...
- morelisp 4y agoThese are largely (if not completely) applications of existing zero-knowledge algorithms to blockchain data, not the application of blockchains to solve some difficult ZK problems or make a useful-outside-of-blockchain novel ZK construction.
- bowsamic 4y agoAnd? Applications of research brings new insights into that thing.
- stale2002 4y ago> The pressure to keep the chain consistent and unified is a purely social one So then the innovation of cryptocurrency was an economic one. It does have the word "currency" in it, so that should not be surprising.
- morelisp 4y agoI'll leave the question of whether it's economically interesting to economists and sociologists (though I suspect the answer is it's not at least in this regard, as the pressure to use the same non-blockchain currency seems not too different across the sweep of history). The claim was: > It turns out that "distributed linked list" is actually a difficult problem that involves very interesting challenges It's not that.