Adrian Cole

I zipkin java we have an httpsampler (due to popular demand) which happens before a trace ID is selected. The side effect is basically it sets the sampled bit (which...

note the service name in jaeger/opentracing is something inherited from zipkin, and while their definition may have changed since it is important to note the service represents the "current process"...

yep in a lot of our instrumentation, names are resetable until late. This is because often a better span name isn't available, yet. For example, the route name of an...

Fyi in finagle there is a "fragment sent" and received annotation for this. It doesn't differentiate as to why one or the other. Importantly, whatever we have in this section...

PS while there isn't a lot of instrumentation on the java side, it might be unclear what this buys us. The thing is most libraries know authoritatively what they are,...

I don't want to go too far into the details of Kafka, except certainly what you mention is not possible now ( to instrument end to end ) Main thing...

besides batching the other thing that will be a problem if trying to model at the lowest abstraction (messaging driver) is fragmentation. For example, rabbitmq will frag a message over...

this was added in zipkin to accompany the http.route tag change. if helpful I can pull up the discussions. on export we likely would rewrite the name to include method...

besides pre-existence in stackdriver, can you enumerate a real-world use case, like what an existing customer does with string types?

@lychung83 thanks for the info. I would recommend adding this rationale to the description!