jim, i've done my best, i really tried, but
i just can't see a use-case for your pgdiff.

i did rework some of its output, to show
how my tool could successfully aid yours,
and i am willing to show that, if you want.

but i was starting to get the feeling that
i was just turning _your_ output into the
output generated by _my_ tool, and how
does it help us to have identical output?

you've said pgdiff is intended for doing
comparisons between different editions,
and not ones aimed at finding the small
differences that happen during proofing.

but, as far as i can see, your output gets
mired even on the smaller changes, so i
can't imagine it's better on bigger ones.

just as one example, there's this:
>   { [Illustration: } { The } { Bakery } { and }
>   { Shops } { of } { the } { Russians--Later }
>   { the | property. }

that's a line of text from one of the two
comparison files, which got chopped into
its individual words.  i could only imagine
the mess if a whole paragraph or section
was added in the second comparison file.

so, sadly, i just can't find a use-case here.

but, you know, if it scratches _your_ itch,
that's all that really matters with a tool...

-bowerbird

p.s.  however, i would really like to hear
from anyone who employs jim's tool and
finds that it helps them with their goals,
because if the failure to see a use-case
is _my_ failure, i want to know about it.