Yaacov Rydzinski

@AnthonyMDev thanks for the comments! I think you are right on. I also have some changes coming in to the validation edits to reflect how the PR came together. namely,...

Comments from WG: Unions don't *implement* interfaces, their member types do. This keyword may be helpful to create a contract by which the union declares that its member types all...

In terms of the first point, I imagine we could think along the lines of a new intersection type that would operate on Union and Interfaces types. It would contain...

@leebyron does that feel like the correct precedent from other type systems?

I think the latter is also done now but I may have missed a case, will have to double check, if you can point to more areas to update, of...

But for completeness, I think the question is whether you should be able to do ... on AorB

> In terms of the first point, I imagine we could think along the lines of a new intersection type that would operate on Union and Interfaces types. It would...

spec PR: https://github.com/graphql/graphql-spec/pull/941 discussion @ https://github.com/graphql/graphql-wg/discussions/944 implementation PR: https://github.com/graphql/graphql-js/pull/3550 The main practical difference after further reflection seems to be how to handle adding a new type to the union that...

Please see discussion linked above. It has the real world use case with unions implementing node linked above. I think if we think of a third solution it makes sense...

*Decision Record* We decided [at the last WG](https://github.com/graphql/graphql-wg/blob/main/agendas/2022/2022-05-05.md) that if unions implement interfaces, then they have directly querable fields. This PR and [the implementation PR](https://github.com/graphql/graphql-js/pull/3527) have been updated with the...