Is a "generic solution" for not duplicating spans still under construction? Is there any reference

[svrnm]

On the page on native library instrumentation in the section Nested network and other spans the following is stated

WARNING: Generic solution to avoid duplication is under construction 🚧.

This has been there since #783, so I am wondering if this is resolved and we can update the documentation accordingly or if this is still under construction and if it is still under construction if there is a spec enhancement or OTEP that we can reference for that? @lmolkova @SergeyKanzhelev @austinlparker, sorry to ping you on this old matter but maybe someone of you can clarify?

[lmolkova]

It hasn't moved much since that.

You can find the history in open-telemetry/oteps#172 and mostly open-telemetry/opentelemetry-specification#1767 (and other issues mentioned in the OTEP).

TL;DR: each language has its own custom way. Java uses the proposal in the OTEP, JS, .NET and Python have APIs to suppress nested HTTP instrumentation on the scope of this call (sometimes they are internal and used within otel libs).

From the article's standpoint, we can either say that there is no generic solution, mention a few language-specific ones, leave it as is, or just say that native instrumentation should be cautious about it.

[svrnm]

Thanks for the clarification, I'll think about a proper wording and update that accordingly.