Roman Ivantsov

> The main practical difference after further reflection seems to be how to handle adding a new type to the union that does not implement the interface. Using intersections, it's...

@yaacovCR - I do not understand anything. In my understanding: as soon as two same-name fields but of different types (your 'friendly') from different interfaces collide in a single type...

that looks really problematic to me. You suggest essentially a syntactic sugar, a shortcut that allows you to skip explicitly declaring a type with fields. Benefits are tiny I think....

Note - as other language I use c# as 'other language' that has tuples and maps (dictionaries). > Tuples and maps are a necessary feature and are supported in most...

@benji the commentary text does not matter, the examples should be CORRECT GraphQL docs, on their own. Unless of course you show a mistake. Again, my reasoning - we are...

What I really do not understand is this RESISTANCE to this simple change. If we add 'mutation' - will it change the story you are telling here, about executing selection...

@michaelstaib again, why adding notes which might confuse even more, rather than simply adding 'mutation' and ending all confusions? Mutation sets are 'selection sets', they return data which can be...

Guys, I do not mind an example showing 'serial execution of selection sets', for whatever reason the SERIAL execution is selected. Service is free to choose any way. It is...

@yaacovCR , not sure what you mean: > the spec as currently defined only has one serial execution the spec dictates SERIAL for mutation, but for queries it leaves the...

> The text as currently stated in the spec is, however, correct. so, "scalar values must not differ" - it is correct? which values? we are at early validation stage,...