[r-t] 8 spliced atw 7com

[Mark Davies]

Alan writes,

> But isn't that a bit of a failure of the restrictive nature of a
> "conventional composition of spliced" if compositions finishing at the snap
> cannot be considered round blocks?

You may certainly consider a snap-finish as a round block. What do you 
get if you do? The resulting arrangement can be described in two ways:

1. A composition containing a partial lead of (one of) your method(s).

2. A composition with an additional method, the additional method being 
two changes long (in the case of a backstroke snap finish).

These are two valid ways of describing the same composition. In the 
latter case, we still have a "conventional" composition, just with an 
extra method spliced into it. The argument about COM is unchanged. For 
instance, the touch YYYY-L LYs(Y) becomes a composition of three-spliced 
with 5 com:

  Y->L
  L->Y
  Y->SY
  SY->Y

(Where "SY" is a new method consisting of the first two changes of 
Yorkshire). We could happily start the composition at the second 
course-end, ringing the 2-change lead of the "SY" method and then 
changing into Yorkshire.)

In the former case, we have to decide whether a "restart" of a lead of a 
method, without changing the method, is a change of method or not. 
Clearly the conductor has to make a call, so he may view it as a COM. 
But from the composer's point of view, I think it is not really a change 
of method. To see why, take the same touch as above: by my argument, 
viewing it as a round-block with lead-restarts permitted, this now has 
just 2 com:

  Y->L
  L->Y

If we start at the second course end it is still just 2 com:

  L->Y
  Y->L

This is because the "lead restart" of Yorkshire is not counted as a com 
(the method stays the same). But here's the thing - doing it this way 
means the composition can be started from any change, even mid-lead. You 
could for instance start at the half-lead of one of the London leads, 
and still ring the same touch, with the same COM. The fact that you ring 
two partial leads of London has no effect on this method of counting 
COM, and that is I think why it is the best approach.

(Note that the modified touch LY-L LYs(Y), treated as a round-block with 
lead restarts, gives 4 COM by this counting method, exactly as you'd 
expect for a "normal" round block.)

So that is how the method of counting COM can be generalised to round 
blocks containing partial leads. Very straightforward really (although, 
of course, not currently "allowed" under Methods Committee rules).

But overall, for "conventional" snap finishes, it is generally more 
sensible to say:

3. That we do not have a round block with respect to the leads of the 
methods included in the composition, and hence that the start and finish 
do not join up.

MBD