[r-t] Shades of truth

[Richard Smith]

Don Morrison wrote:

> Is "round block" really the term we want to use here? Is it not
> normally used to describe a block that not only starts and ends with
> rounds, but starts and ends at the same point of lead? So, for
> example, Brian Price's 5,090 of Cambridge Major is not normally
> considered a round block, right?

No, 'round block' probably wasn't a good choice of term. 
But equally, I think 'round block' can be ambiguous, meaning 
both a block that starts and ends at the same point of the 
lead, and also a block that starts and ends at the same row. 
Which is usually obvious from context.  But yes, it would be 
better to use a different term.

In fact, I suspect the choice of terminology is a major 
challenge to this aproach.  That's largely why I used 
single-letter abbreviations.  But I think an actual proposal 
would need names for them all.

>>  [U] includes every row at most once; and
>>  [R] starts and ends with the same row.
> ...
>>  [C] includes every possible row an equal non-zero number
>>      of times.
> ...
>>  [S] is completely divisible into one or more
>>      non-overlapping blocks consecutive rows such that each
>>      block individually satisfies [U], [R] and [C].
> ...
>>  [B] is completely divisible into one or more
>>      non-overlapping blocks consecutive rows such that each
>>      block individually satisfies [U] and [C].
>
> I suspect this is not the only formulation we could start from, and
> arrive at versions equally sympathetic to existing practice, but
> implying different relaxations going forward. That is, how we choose
> to be descriptive may, in practice, subtly lead to prescriptions.

I can certainly believe that it /may/ do.  But I would be 
curious as to whether an uncontrived proposal actually did.

I think my [R] criterion is sufficiently distinct from the 
others that it would be hard to formulate it differently. 
And I think that, on historical grounds, if nothing else, we 
can justify something similar to the [S] => [B] => [C] 
series, even if not exactly how I have it.  (I'm comfortable
using history to justify one classification over another if 
there are no technical reasons to prefer one.)

I suspect the biggest issue is how you divide up the 
combination of [U] and [C].  My approach was to basically to 
write one rule for major and above (which is [U]) and 
another rule for triples and below (which is [C]).

But I'd be interested to explore some alternatives so see 
whether they actually do produce noticably different 
results.

>  "We should prescribe the minimum that we need to call it 'ringing', for
>   example what changes, rows and truth are, and just describe the rest."
>
> Why do we include truth in that list?

I could easily imagine I might have written something like 
that, and I think the reason is simply that I hadn't thought 
about it enough.  I've thought about methods and 
compositions a lot so I knew that they were complicated. 
But I hadn't really given much thought to truth as such.

> I'm certainly happy to have it be a thing that varies with 
> time and purpose. I suspect some others may be more 
> uncomfortable with that.

I'd be interested to hear from those who are unconfortable 
with the idea.  Yes, I can guess what some of the objections 
might be, but I would like to actually hear them.  Last year 
Martin Cansdale said "I don't think the solution for this is 
to recognise as a peal [...] any old rubbish."  And of 
course the problem with this is that we almost certainly 
cannot unanimously agree on what constitutes "any old 
rubbish".  By having multiple standards of truth we're 
recognising that people disagree and allowing people to see 
immediately what standard of truth applies to what 
performance.

RAS