[r-t] Methods Committee proposed change 3.0(D)

[Don Morrison]

On Wed, Mar 22, 2017 at 4:31 AM, Mark Davies <mark at snowtiger.net> wrote:
> Ah! Read it again... it doesn't say that any more.

You may not have intended it to say that, but if it doesn't it sure is
subtle and, I think, ambiguous:

> "16. A peal shall comprise true changes. This means that none of the
> possible rows for each effective stage shall appear more than once
> more than any other of the possible rows. The effective stages for
> different categories of peals are defined in section 5 below."

Note that a peal "comprises" true changes (by the way, I think "true
changes" is an horrible term, since it's the rows we're comparing). It's
hard to figure out how that cannot mean that each of the rows of the whole
peal must not appear N or N+1 times. It does not say that the peal is
partitioned into blocks, which may be of two different categories, the rows
in each category being tested independently. But even if you're convinced
that it does say that they are so partitioned, then when you do have two
categories of component blocks each may independently have a partial extent
included in it, which I strongly doubt is your intention (though I'd
personally have no objection to it, but believe myself to be in a small
minority on that point).


-- 
Don Morrison <dfm at ringing.org>
"Serendipity is a shortcut to joy."
        -- Eli Pariser, _The Filter Bubble_
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ringingworld.co.uk/pipermail/ringing-theory/attachments/20170322/b764fe20/attachment-0001.html>