Mark Larah

> state in the spec that every non-metafield schema element should have exactly one coordinate Done in https://github.com/graphql/graphql-spec/pull/794/commits/62ce19aafb0e0be105cd248e3e044d3f5103a897

Thanks for the feedback! Ooh, I like "field coordinates", that's slightly more descriptive and less awkward than "type/field pair". Not 100% sure on the process here, but I'll send a...

fwiw, assuming this functionality doesn't yet exist, i could imagine some fancy build-time sourcemap-like solution that looks at the ast and stores the ordering of relationships between sibling function calls....

Thanks @cartermp! Would be really curious how existing UIs calculate the critical path - the current data provided in spans doesn't seem sufficient to do so accurately as we don't...

@tedsuo > @magicmark can you describe a situation where sibling spans are blocking each other? Consider the following (synchronous) code: ```python @trace def foo(): bar() baz() qux() ``` Assume that...

Sorry....I think i take this all back actually! As has been pointed out, jaeger shows the critical path (marked by the black line) out of the box when I download...

That said....this is an interesting trace which I'll dig into more I changed the code such that foo7 actually blocks on foo3, but it still shows that it's blocking on...

@cartermp thanks! So I haven't fully dug into it yet, but I'm wondering if i'm just running into what I originally suspected which is that the trace data fundamentally doesn't...

To confirm, this is my minimal repro: This visualization is wrong ^ `/get foo3` is **not** blocked by `/get foo2` Produced by this code: ```python async def make_fake_network_request(resource_name, length): with...

@austinlparker yes and no - in our current real life situation, yes, we're mostly focused on the network calls emitted by one particular service. But we might in future also...