Thank you for your comment. You are indeed correct. It unfortunately also means that current programs using default(none) would fail, since it is no longer predetermined. The ARB usually tries to make sure that this doesn't happen, so we may have to revisit this to see whether it should be like a loop iteration variable (i.e., predetermined but allowed to be specified) so that current programs can continue to work. In any case, I have put your comment on the list of items to be reviewed before the specification is finalized.