[r-t] Anything Goes vs Peals Mean Something

[Mark Davies]

Simon Humphreys writes,

> About 40320 r-t messages ago I asked where do you draw the line for the
> number of different calls allowed in a peal.  The answer (paraphrased) was
> that a line doesn't need to be drawn at all, in this case, and not in many
> (most?) other cases also.

Some things are essential to the definition of a peal, some aren't. Peals
are:

1. True.
2. Of a minimum length.
3. Composed of changeringing methods.
4. Of a reasonable quality of performance.

No-one is arguing over that, are they? What we're discussing is questions of
degree in all four.

My point is if you weaken these four standards much beyond what we have now,
you devalue the meaning of "peal".

Yes, it would be nice to be able to ring partial extents, so broadening
definition (1) slightly from what we have now; but we don't want to dilute
proof completely, we don't want "obviously" false stuff being rung (hence my
current discussion with RAS). It would DEFINITELY be nice to be able to ring
more interesting methods, classified in a more sensible way; hence
broadening definition (3) really  does need to be thought about. This is
where most of the work lies, and where we should probably be focussing our
attentions.

But I can't really see why we're too worried about (2) and (4). The existing
definitions are fine. Allowing people to ring an extent of Triples less a
few changes sounds really stupid to me, all for the sake of some pointless
rationalisation.

> 5040 is mathematically neat and satisfying, but having two standards is
> not.

Yes it is, absolutely! It is beautiful. If you simplify too far, you lose
that beauty. And you are forced to adopt the arbitrary length - 5000 - which
is meaningless to our art.

Not everything has to be reduced to an atom. Apply your sledgehammer 
elsewhere!

MBD