A way to stop the "overriding" may be to include "sender must be
clear" in the *definition* of by announcement rather than as an additional
property. It's awkward to say "you did that by announcement, but it failed
for reason B" and better to be able to say "you didn't do it by announcement
because the announcement was ambiguous" (the same way you would do if
other parts of the announcement were unclear). (I think doing so would
fix what you're seeing in R2141).
On Thu, 14 Jun 2018, Aris Merchant wrote:
> On Thu, Jun 14, 2018 at 1:03 PM Kerim Aydin <ke...@u.washington.edu> wrote:
> >
> >
> >
> > On Thu, 14 Jun 2018, Aris Merchant wrote:
> > > Change the power of Rule 478, "Fora", to 3.1.
> > >
> > > [I'm astonished that no one has thought to do this before now, given
> > > that this rule contains conditions for ALL actions taken by sending
> > > messages
> > > to work.]
> >
> > Unless there's a specific conflict, I don't see the need. And are you
> > sure (i.e. done due diligence) that there's no "rules to the contrary
> > notwithstanding" in a different power-3 rule that needs to overrule this
> > rule in order to work?
>
> I have now. There are none that shouldn't be overridden, and at least
> one that should (Rule 2141). The point is that this is a very easy
> restriction to override. Any power 3.0 "X CAN Y by announcement, rules
> to the contrary notwithstanding" would bypass the restriction, despite
> the fact that it clearly isn't trying to. I'm not sure that it's
> absolutely critical that the power of the rule be increased, and I
> know we're very reluctant to increase anything above 3.0. However, it
> seems to me that R478 is exactly the kind of critical infrastructure
> that shouldn't ever be overridden, it being one of our most basic and
> most ancient rules.
>
> -Aris
>
