[r-t] Proportion of Surprise Methods

[Don Morrison]

2009/3/20 Richard Smith <richard at ex-parrot.com>:
> Actually, they're not: they're contradictory.

I suspect they probably are, but I don't think this example
demonstrates any contradiction.

As I read (E)C.2, it means that to be surprise there must be an
internal place for each of the principal hunt bells at each of its own
cross sections. That 18 place notation at what is the lead end for one
of the hunt bells is, according to my understanding, a cross section
for the other hunt bell.

And thus I believe this method is unambiguously a delight method,
according to the current Decisions, a property I think is invariant
under rotation, though I'd not want to have to prove it.

And having said that, I'm pleased to confirm that if you plug either
place notation into

  <http://ringing.org/main/pages/method-search>

it also believes them both to be delight methods. It's been a while,
but I think I did try to think this through a bit when writing that
code, and it appears I got it right, or at least consistent, in at
least this case.

I'm not trying to argue that this is necessarily a good definition of
method class, and certainly not one that appears to have any practical
utility. But I think, at least with respect to this example, it is at
least self-consistent, and also does have the virtue to reducing to
common practice for the common cases.

Or am I, as usual, missing something?



-- 
Don Morrison <dfm at ringing.org>
"The press was to serve the governed, not the governors. The
Government's power to censor the press was abolished so that
the press would remain forever free to censure the Government."
     -- Hugo Black, decision in New York Times v. United States