Paper: CRDTs

[mmcgrana]

Convergent and Commutative Replicated Data Types (Shapiro et al.)

From #40.

[ferd]

This is a pretty good paper -- I've always liked it -- but I feel it's very specific to distributed systems theory and is way closer to computer science than engineering.

It doesn't show many of the tradeoffs and risks (infinite growth, or need for hand-off counters), and while it's one of my favorite papers (really, data structures that remain conflict-free and always available during and after netsplits are pretty damn sweet), I feel it might not fit very well into 'services engineering' or cloud infrastructures?

[mmcgrana]

@ferd I generally agree. I'm going to keep this issue open and noodle on it some more but doesn't seem like an obvious candidate right now.