Philippe Marschall

Just do let you know, finalization will likely be deprecated for removal, see [JEP 421](https://openjdk.java.net/jeps/421).

I would not support `LocalDateTime`. `LocalDateTime` is not an instant and therefore does not identify a point in time. You would need a time zone for that. Which means you...

@RockyMM sure, `Instant`, `OffsetDateTime` and `ZonedDateTime` are a great addition and completely fine.

The JDK is stock apart from disabling TLS 1.0 and 1.1 and changing the entropy gathering device. AES256-SHA256, AES256-SHA, AES128-SHA256 and AES128-SHA are all enabled. This is the list of...

I am reasonably sure the client sends 0x003C. Here's an excerpt from the Java TLS debug log of the client ``` ClientHello.java:653|Produced ClientHello handshake message ( "ClientHello": { "client version"...

The communication was not wiresharked instead we used [Java TLS debug logging](https://docs.oracle.com/en/java/javase/11/security/java-secure-socket-extension-jsse-reference-guide.html#GUID-31B7E142-B874-46E9-8DD0-4E18EC0EB2CF). We believe the 1023 bit DH key is a bug / misconfiguration on the server side, even it...

No, not at all. It’s just a clean up or code quality thing that affects only the implementation. The behavior users see is exactly the same.

I thought the same thing. I found that it's in general easier to put JMH benchmarks in a dedicated Maven submodule instead of fitting them into an existing one. I...

I understand that changing the ABI may not be an option. What would be nice if we could formalise / specify the behaviour for implementations with a fixed scale.

#119 removes only one part, more remains to do Removing `AbstractQuery#KEY_QUERY_TIMESTAMP` is a breaking change, I'm not sure we want to do this, maybe only deprecate. As for the BP...