Nik Clayton

Been thinking about this a bit, and poking at the code. Trying to think about the best way of representing this in the low level Flag type in `state.rs`. Couple...

Having read that doc, I think this could be done in a way that doesn't violate the stability guarantees (as defined in https://github.com/open-telemetry/opentelemetry-specification/blob/main/specification/versioning-and-stability.md#stable), in particular, in a way that doesn't...

The CI failures do not appear to be related

First, here's what Jaeger does with my service without this PR. In this example the trace spans two different processes that are part of the same service, and they are...

This is the same thing, but with this PR.  The first result shows two entries for the same service, and includes the namespace (empty...

The service is `replica`, and each instance of that service is reporting a different `service.instance.id` tag value (they don't have a `service.namespace`). These fields are from the OpenTelemetry resource semantic...

The code's the simplest thing to prove the concept, but isn't production ready -- some code is duplicated, some information is passed lower down the call tree than I'd like,...

> You seem to be making color decisions by default, based on a specific tag that may not be applicable to every user. Sort of, but in exactly the same...

> The properties you're considering are not standard, nor produced by many instrumentations. That's true -- they're an experimental recommendation from a sibling CNCF project. I'm coming from the perspective...

@tiffon > OTOH, we now have the ability to specify the UI configuration via JaveScript, Did you actually mean JavaScript there, or just JSON? > That's a large scope. Is...