Robert Pająk

> Jack's proposal during the SIG meeting of emitting one array field instead of multiple fields sounds reasonable to me. This would cause performance overhead as well. > Sumo Logic...

Even if the change would be seen as not acceptable for Collector: - other exporters (e.g. console exporter) can take benefit from it - OTLP exporters could have a configuration...

> This means that it's acceptable for an implementation to assume the keys are unique and reject a payload with duplicate keys as invalid. I agree. I updated https://github.com/open-telemetry/opentelemetry-proto/pull/533 to...

I changed https://github.com/open-telemetry/opentelemetry-specification/issues/3931 to draft as changes in OTLP are undesired. Still, updating the definition of on model level can be helpful as it would postpone the deduplication logic to...

I opened https://github.com/open-telemetry/opentelemetry-specification/pull/3938

> I am not sure if it will be good idea to support only managed code on arm64. It should be done together with #1865 I think it is OK....

> @lachmatt I missed something earlier - I think this would cause an issue due to difference in versions. https://learn.microsoft.com/en-us/dotnet/api/system.uri.unescapedatastring?view=net-8.0#remarks > > Consider a service using old sdk injecting baggage...

@open-telemetry/dotnet-maintainers This bug makes the baggage propagation not working properly with other SDKs.

@open-telemetry/dotnet-approvers PTAL. We would like to have it fixed at least for 1.8.0.

@tigrannajaryan PTAL. I addressed your comment.