Gerhard Stöbich

Tried with follwoing versions and I see no crash: ``` "@opentelemetry/instrumentation": "0.30.0", "@opentelemetry/instrumentation-http": "0.30.0", "@opentelemetry/sdk-trace-base": "1.4.0", "@opentelemetry/sdk-trace-node": "1.4.0", "superagent": "8.0.0" ```

Actually there is no single right way. It's up to the user to decide which variant to use. Some points one has to think about (there are likely more points):...

> So when starting a new span you can specify a context and in that context there might be a parent span and if so that becomes the parent of...

I was not that involved in the spec discussions regarding this. As far as I know the main reason was that ontext is more generic and it can be used...

regarding always return a span: How and a user differentiate between a non sampled trace (which is represented by Noop/NonRecording/... spans) and no trace active at all? This is relevant...

Well for propagators it's fine but I still don't see the advantage to create a Noop instance just to return something. Or to add more APIs on span like (IsSampled(),...

if you call a non existing method on a Noop it will also throw. So well, wrong usage results in undefined behavior - as one would expect. Maybe a bit...

> * SDKs MAY log a warning if an invalid/noop span was returned. I think we should not issue a warning as it is perfectly fine that no trace is...

I don't think it's fixed as abstract `init` is still called from base class constructor [here](https://github.com/open-telemetry/opentelemetry-js/blob/1a0e0c44d2d5930a9c140f3ad650da04b43784d3/experimental/packages/opentelemetry-instrumentation/src/platform/node/instrumentation.ts#L42)

According to spec resource is immutable. I think delaying the read until first use in e.g. exporter sounds to me like it's still spec conform. I don't think the `await`...