PtrMan

I have to revert the stupid package-info.java

What about documenting it with literate programming? The only downside is that it's hard to edit the code.

New lines can improve the understanding of the code - I see it as a way to seperate microtasks of a method which don't deserve own methods.

The naming scheme of methods to check/examine seems to be inconsistent to me. ex: indepVarUsedInvalid() (examines if the indep var is used in a invalid way - this is not...

Why is this strange? One piece of the code calculates the budget, the other the truth. I don't see the problem here. The other piece of the code is special...

4. as I tried to explain it in the IRC - split the deriver work into n sets for n threads. This works well with Fork-Join and is the most...

I don't see your problem each thread does work with different inference rules for the same premises. One thread could compute deduction for inheritance while the other computes similarity. Just...

4. is neither a pipeline nor a stream approach because it all will be shortlived and joined together after the inference is done (for each step).

This was just a example - we have like 600 rules of different families in narjure - there is enough parallelism to feed a few cores

http://maven.apache.org/surefire/maven-surefire-plugin/examples/inclusion-exclusion.html