[r-t] applicability and timing (was The null change)

[Don Morrison]

On Tue, Dec 30, 2014 at 8:32 PM, Matthew Frye <matthew at frye.org.uk> wrote:
> [regarding the plain course of a non-β method] all of the sequences
> obtained by applying the method to the N! possible starting rows.

I'm sorry, I don't know what this means. Do you mean the union of all
the rows produced by starting from each of the possible rows? That
seems exactly the opposite of a plain course, and, except for deeply
pathological methods that don't work on certain rows, would be the
extent. Or do you mean a non-β method can have multiple, different
plain courses? Or something else? Or am I just confused and completely
missing something?

>  I might like something along the lines of "the method definition
>  must make every bell to move at least once" as the only sane
>  starting point (I think trying to discuss "courses" or "leads" in
>  the context of non-β methods is a lost cause), but that needs some
>  serious work. Apart from anything else, there's nothing to stop
>  someone from adding in an obscure rule that would only be engaged
>  by a row that will never be rung to artificially satisfy that
>  requirement.

But if some idiot wants to do something stupid, there's nothing we can
do to stop him. He'll find another way around it with a call or
something. There's no way we can stop idiots from doing stupid things,
if they are determined to do them: they're so much smarter than we
are. And it's not are job to stop them, anyway: description, not
prescription.



-- 
Don Morrison <dfm at ringing.org>
"Never underestimate the power of human stupidity."
          -- Robert Heinlein, _Time Enough for Love_