4 ms·

There's a better rebuttal(*) of CAP in Kleppmann's DDIA, under the title, "The unhelpful CAP theorem". I won't plagiarize his text, instead the chapter referen

There's a better rebuttal(*) of CAP in Kleppmann's DDIA, under the title, "The unhelpful CAP theorem". I won't plagiarize his text, instead the chapter references his blogpost, "Please stop calling databases CP or AP": https://martin.kleppmann.com/2015/05/11/please-stop-calling-databases-cp-or-ap.html https://martin.kleppmann.com/2015/05/11/please-stop-calling-... (*): rebuttal I think is the wrong word, but I couldn't think of better.

cowboylowrez 2y ago
Really good read, in fact all of these discussions are really great. I liked the linearizable explanation, like when Alice knows the tournament outcome but Bob doesn't. A super extension to this would underscore how important this is, and the danger of Alice knowing the outcome but Bob not knowing, just extend the website a bit and make it a sports gambling website. A system under such a parition would allow Alice to make a "sure thing" bet against Bob, so the constraint should be that when Bob's view is stale, it does not take bets. But how does Bobs view know its stale? It has to query Alices view! Lots of mind games to play!