After mulling it over, I reckon while it might be handy to have ACL
stuff in the same namespace(s) as SIOC, I don't see their absence as a
problem - such stuff is likely to live behind a firewall anyway, so it
doesn't matter whether it's a text file, RDF or whatever, probably can
be left to per-case requirements.
The only thing that really tripped me up with SIOC is the plethora of
inverses, too late to change in the spec, though maybe the docs could
do with a bunch more usage examples.
Sorry to sound even slightly negative, it's only because SIOC is 99%
on the ball.
Cheers,
Danny.