Using must is a no.
This rule should not be absolute.
You can advise people to not put a variable, but variables may need to happen from time to time.
The WWW variable is there so make(1) can create things out of it.
The ports tree is to be used by developpers, not end users. End users have packages and should not have to interact with the ports tree ever.
If a developper needs to know the actual value, they can run make -V WWW, we could even add a few targets, like make open-www that would do stuff like xdg-open "${WWW}".

As the sentence says "should", it's an advice, which means, it's not enforcable, so, it's just more text that do not really add any no value.

In D30315#1225899, @mat wrote:

Also, ansible ahould probably not be using USE_PYTHON=concurrent

I'd rather we used ?= here.

With the s/SUDO_DEPEND/SUDO_DEPENDS/ that Mateusz talked about, feel free to push it.

The only reason a port will need sudo during build is to "detect" that it exists and work with it when being installed.
There will never be a port that only needs sudo during its build when it's not needed during its runtime.
A port must be buildable as a user, so sudo is of absolutely no use, also, ports are built as nobody, and nobody is not in sudoers.

Another thing, only the build param is needed, the run parameter should not exist, and the RUN_DEPENDS should always be added.

In D51165#1189817, @arrowd wrote:

Are you fine with p.5? Zig distfiles get extracted twice at the moment. This can be fixed with introducing an opposite of EXTRACT_ONLY called EXTRACT_EXCEPT or EXTRACT_NOT. Would that be acceptable?

This looks fine. Maybe put the ZIG_TUPLE variable in a Makefile.tuple, and .sinclude it as it's done for cargo.