[r-t] Pick and mix [was Candidate definition #10]

[Mark Davies]

RAS writes,

> You cannot produce a set of definitions that permit all of "good stuff"
> while disallowing all "rubbish".  The line between "good stuff" and
> "rubbish" is simply too subjective.

Yes but Don's #10 disallows lots of good stuff. A lot more than my
definitions, which you slated for not allowing stuff. :-)

Anyway, if we accept the distinction between good and rubbish is somewhat
subjective, we should still explore what is the best compromise. To aid in
this I have created (a first draft of) a "pick and mix" definition of the
truth of a peal. Basically I've listed several rule variants, and you can
pick which ones you like to create a wide range of definitions from VERY
STRICT to ANYTHING GOES.

For example, the strictest rule set is (A, J) only. That would mean that you
can only ring peals on one stage, every method has to be of that stage,
every change has to be unique, and you can't ring more than an extent. Very
draconian.

You could relax the rules to something like the current CC position by
adopting the set (B, F, G, J) with X (the number of stages) = 2. This allows
multiple-extent single-stage peals, but no partial extents; it also allows
variable cover proved at the higher stage, and peals of e.g. Doubles and
Minor in whole extents.

You could relax everything completely and pick the set (E, I-E-3, M) with
X="multiple". This allows you to do anything - any number of stages, no care
in the world about repeated changes. Surely that's a complete waste of paper
though.

Alternatively you could achieve my proposed definitions with the set
(D, I-D-2, M) with X=2. This is *almost* as relaxed as you can be without
resorting to "anything goes".

Or alternatively alternatively you could pick a different set to achieve
whatever you want. What do you like Richard? (If some option is not present
in the pick-and-mix set, please add it.)

(I'll put the actual pick-and-mix ruleset in the next mail, 'cos this one is 
already too long).

MBD