Phillip Krüger

I agree with this. Is this in the spec though ? If not we might want to add that rule there too.

@t1 - Ok, So if this object is also use for an output, should we remove the (already there) Input ? If not we will have an error as both...

Maybe in this case, the user should just use `@Name` to correct the name ?

Why would you use ***Input as a classname ? We currently would control what is input and what is output via annotations on a normal class.

@t1 - can you look at this ?

@t1 @andymc12 @jmartisk w.d.y.t of the above idea ?

So the idea is as follows, since we build the schema anyway, and know all the queries that use a certain source method, we do the following: if the root...

For example, using the GitHub API: ``` { search(query: "...bla bla bla", type: ISSUE, last: 100) { edges { node { ... on PullRequest { number title repository { nameWithOwner...

https://docs.github.com/en/graphql/overview/explorer

You mean in this case you should just use the Dynamic client ? I don't think that is ok. We should be able to model this