Re: Order of static pattern rules is significant while IMHO it should not be

> I think static pattern rules without recipes are rare (at least I've
> never seen or used one)

I could quickly lay my hands on what looks like a handful:

./subrules.make:384:$(SOFTWARE)$(PROJPATH)/%: %;

./subrules.make:392:$(FLEX_GEN_CPP): %.gen.cpp: generated/regenerated.%
./subrules.make:393:$(FLEX_GEN_H): %.gen.h: generated/regenerated.%

./subrules.make:530:clean clobber: %: %.generated

./trules.make:642:$(ORPHAN_OBJS): $(TARGET_DIR)/%.o: remove_orphans.%

The first one looks to be trying to persuade Make about the equivalence of a certain sort of relative path and an absolute one.  The rest seem to be more directly code to tell Make about side-effects - build this first and then maybe what you wanted will have magically appeared, or disappeared.