Benjie

> I think this depends on if c is nullable or not (with b propagating the error to a). That's right, yes - exactly as you describe. > Which means...

Plan from today's WG: 1. open a new PR that clarifies "response position" without renaming to "execution position" (which could be ambiguous w.r.t. visiting fragments?) 2. rebase this change on...

Indeed; we've discussed having a backlog for tracking things that need forward porting. Essentially the `16.x.x` branch just needs to be merged into `main` periodically. We considered changing v17 to...

@leebyron I think this matches what you had in mind? (Leaving the resolvers of the introspection system to be the behaviour of those defined within that section of the spec.)

That's weird... CI didn't run. Closing and reopening.

All sorted. Had to fix the codecov setup.

Thanks for the submission; we will be unable to fully review it until you have agreed to the contributor licensing agreement - please follow the instructions above. Further, please also...

I think there’s some support for this position already; would you like to step up and be the champion of it? Read about our contribution process here: https://github.com/graphql/graphql-spec/blob/main/CONTRIBUTING.md

On a related note, I think we're going to recommend that the URLs be `/graphql/persisted//` - see https://github.com/graphql/graphql-over-http/pull/305/files#r1706576324 As such, overlapping the DOC_ID and OPERATION_NAME may be a smidge redundant;...

@magicmark It's not merged yet, but the first draft is in https://github.com/graphql/graphql-over-http/pull/264 which is then extended by https://github.com/graphql/graphql-over-http/pull/305