@bascule Nice post, although SCP is dead on arrival. It has no real mechanism to prevent forks. Very dangerous.
-
-
-
@bascule Forks seem likely, and once they occur reconciliation seems impossible. -
@taoeffect Bitcoin has exactly that problem (losing accepted writes/transactions in forks). At least SCP requires a quorum to accept a write -
@bascule Bitcoin does not have the problem I am describing at all. Bitcoin has a way of defining "truth", whereas SCP does not. -
@taoeffect Bitcoin loses acknowledged writes. From a CAP theoretic perspective, it sacrifices partition tolerance:http://codahale.com/you-cant-sacrifice-partition-tolerance/ … -
@bascule@darkuncle Is that surprising? Your link points out that all distributed systems have that property.pic.twitter.com/KL60Qx5J7l
-
@taoeffect that's saying that to tolerate faulty networks (or faulty clients), systems *must* be partition tolerant. Bitcoin isn't. -
@bascule Ah you're right, sorry I misread your tweet. - 8 more replies
New conversation -
-
-
@AKWAnalytics and what "basic economic principles" are those? /cc@pierre_rochard -
Tweet unavailable
-
@AKWAnalytics@pierre_rochard cool story bro
End of conversation
New conversation -
-
-
@bascule Why do u say after partition transactions on losing fork are lost? Won't they be incorporated into new blocks? Except double spend. -
@jiceman no, any transactions in an orphan block or fork are irrecoverably lost -
@bascule Nothing looks at the orphan block to ID tx's not in other blocks and to return those to mempool for inclusion in another block? -
@jiceman no, nor is there a merge algorithm defined. Clients have to replay transactions and assume they’ll be included in subsequent blocks
End of conversation
New conversation -
-
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.