Alex Herbert

I would close it. The javadoc currently reflects what the code is doing. If you read the javadoc as is you should get the understanding that the method is _trying_...

I believe @psteitz is correct that the initialisation of SecureRandom can be slow and potentially blocking. The way it is done in Commons RNG and the JDK source code for...

Using ThreadLocal still has a potential blocking call when a random method is invoked.

Sorry, clarification: It is the _first call_ to ThreadLocal.get() that may block on the instantiation of the SecureRandom (depending on how it is created). After that each thread has a...

I think this change is fine. The reference to Java 1.2 and 1.5 are no longer relevant and adding back a different Java version adds a maintenance burden syncing the...

Re MacOS, I think we have to switch the `setup-java` distribution to zulu. Or just drop the windows and mac os from the actions given the code does not use...

Switching to zulu has fixed the master build.

The title is just from the commit message. The code change is to remove statements that spread the dependency information across site documentation (which is evidently not maintained in sync).