[r-t] Anything Goes vs Peals Mean Something

[Richard Smith]

Mark Davies wrote, in reply to me:

> OK, here's the latest. I don't know, you might like this!
>
[...]
>
> 3. In a two-stage peal, the structure of the composition should make clear
> which changes are treated at which stage.

I take it all back.  I do like this.  It makes it clear that 
we don't like the idea of isolated rows from being picked 
out willy-nilly.  But equally, it doesn't actually proscribe 
it.

So, ignoring the fact that singles and triples are not 
adjacent stages, this doesn't actually disallow the false 
triples example, but it does make it pretty clear that we 
don't like it and probably wish it were disallowed.

Is that a fair summary?

I still don't like your adjacent stages restriction, but 
that's a separate issue.

RAS