Proposal: Newer version-string scheme for the Java SE Platform and the JDK

Stephen Colebourne scolebourne at
Thu Nov 2 22:10:56 UTC 2017

On 2 November 2017 at 21:21, John Rose <john.r.rose at> wrote:
> In that environment, would you rather have a predictable
> lock-step rule for tool development, or would you still prefer
> your scheme, which optimizes for the infrequent "null" change,
> at the cost of guessing on a random variable?

If the JVM/bytecode level has not changed, I can't see any advantages
to anyone outside Oracle/OpenJDK of a version bump.


More information about the jdk-dev mailing list