been finally learning what graphql is due to <%= reasons %> and I have never seen such a clear case of a technical solution to a social problem. that problem being "too many people want to talk to the API devs"
0 Boosts
0 Likes
it seems almost *almost* physically impossible not to write horribly slow data access code with this. their *example code* for getting associated objects is an N+1 query. they acknowledge this elsewhere and go "there exist solutions to this, uh, elsewhere"
this btw is why you should be fairly skeptical about adopting Big Tech Company solutions if you're not a Big Tech Company. they are mostly trying to solve a co-ordination problem caused by their product and team being far too large, and then presenting this as technical sophistication
it sort of reminds of JWT, to be put in the bucket of things that are mostly bad on technical merit, but people adopt them because they're a "standard" and present little friction at a social level
you have to write a whole distinct function for every single use of the field access operator? this is supposed to be easier than doing things the "boring" way?
@jcoglan I’ve long referred to graphql as “a technology designed to avoid a conversation”