[r-t] Are FCH's needed any more?

[Don Morrison]

On Fri, May 21, 2010 at 7:49 PM, Graham John <graham at changeringing.co.uk> wrote:
> FCHs were a marvellous invention to simplify the job of proving compositions
> and find an existing composition that worked for a new method, but the
> question arises  of how much value they have today when a computer can
> generate and check all the individual rows in a composition in a fraction of
> a second?

Certainly FCHs are of less use today then they were in the past. Just
as compositions in whole courses true to all of some summary falseness
are far less rung now than just a decade or two ago.

But I wouldn't say they're as obsolete as a table of logarithms*.

Perhaps I'm just showing my age, but when approaching a new method I
will still generally start by looking at it's summary falseness, and
where that falseness appears. If it's simple enough, often I'll even
figure out at least the in course falseness by inspection of a half
lead rather than using a machine to work it out in detail. Even though
I will be doing some extensive computer searches, I usually find
having some sense of what courses are or are not going to work helpful
for understanding exactly what space to search, or which part heads
are likely to be useful.



* Curiously I just this week consulted a table of logarithms for
  something at work, probably for the first time in thirty years.



-- 
Don Morrison <dfm at ringing.org>
"[I] am rarely happier than when spending an entire day
programming my computer to perform automatically a task it
would otherwise take me a good ten seconds to do by hand."
                     -- Douglas Adams, _Last Chance to See_