Hi, wanted to share this paper that found very interesting https://dl.acm.org/citation.cfm?doid=3178876.3186014

Not certain that I care that " This implies that a server can answer a GraphQL query and send the response byte-by-byte while spending just a constant amount of time between every byte sent." As I read it, they seem to be assuming that the entire GraphQL data structure is fully realized in memory before streaming the response, which is anything but the case!

Maybe I'll read it in more detail later. The interesting/hard part of GraphQL and Lacinia is the async construction of the response data NOT the streaming of that data, as JSON, to the client.

do you know what the reasoning is on the spec 5.4.1.4 “Fragments Must be Used” rule?

Not really, and we don't do a perfect job on that validation, afaik. One theory: they want to prevent clients from "cargo culting" big requests using cut-and-paste, referencing things they don't actually need.

ah, yeah I guess that makes sense