@QuiltProject interested that you mention tuplespaces in blog. Sceptical of ts based on CRDT. I work on Calvin-like https://github.com/vjoel/tupelo
-
@QuiltProject 5. Tuplespace ops are inherently non-commutative. Use a tuplespace when your problem demands coordination.
@QuiltProject 6. When your problem doesn't demand coordination, use coordination-free CRDT or CALM.
-
-
@JoelVanderWerf Thanks very much for engaging! I'll keep an eye on Tupelo, hopefully we can get a chance to chat in person, etc. :-D -
@QuiltProject Sounds great. Best wishes for your talk!
End of conversation
New conversation -
-
-
@JoelVanderWerf 5/ Re: coordination/consensus, that's something that I think can be built *on top of* a suitably-constructed CRDT. -
@QuiltProject Interesting thought. I'll move to the lobsters discussion... https://lobste.rs/s/s6gi7r/distributed_systems_and_the_end_of_the_api … -
@JoelVanderWerf Replied there. Much better venue, as my prior 140 chars on the subject left too much unsaid. :-)
End of conversation
New conversation -
-
-
@JoelVanderWerf 3/ Maybe this means I'm not being faithful to the original principles (esp. of tuplespaces, which I think are overspecified)Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@JoelVanderWerf 4/ But, I'm okay with that, as long as the end result yields a sum greater than its parts.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@JoelVanderWerf 2/ I think both can be usefully characterized as immutable, as per persistent data structures used commonly in FP langs.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@JoelVanderWerf 1/ Right, thanks for that. Everyone assumes that both CRDTs and tuplespaces are necessarily mutable constructs.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.