saurik

Why? What problem did this solve for them? I use Netflix a lot, and I use it on AppleTV where I get automatic updates to the device and the app... the app has never changed except to get slower over the years (and I mean a lot slower). There is absolutely no new functionality they have ever deployed... isn't the app "done"? Is there some serious scaling issue on their backend that is causing the perceived frontend slowness that this will fix? Is this going to massively decrease the costs of the backend? Or is it just that the engineers are all bored and can't accept the idea that the product is either 1) done or 2) needs a new direction (instead of some new tech)?

WorldMaker
It replaced Falcor [1] which was extremely similar to GraphQL but also entirely incompatible, and was owned and maintained almost solely by Netflix staff, with GraphQL which has a large community of developers and users. Released to the public at nearly the same time, after several years in the hands of open source communities the verdict seemed to be: Falcor lost, GraphQL won. In theory this saves Netflix continuing maintenance costs of Falcor (which it can now entirely deprecate/stop work on, though there is no current warning in the repo other than maybe a comment in Issue #1016 [2]), and instead externalize more overall maintenance work to the much larger GraphQL community (including teams like Apollo and at Meta).

[1] https://github.com/Netflix/falcor

[2] https://github.com/Netflix/falcor/issues/1016#issuecomment-1...