jim said:
>   The famous title I reworked
>   still had at least a dozen "obvious errors"
>   that have never been fixed - even after
>   100s of thousands of downloads

one of those times where michael hart
"solicited" my help and got no response
was when he suggested that i could help
on a project aimed at creating an e-text
that was "error-free"...  i rolled my eyes...

first, there's no such thing, as michael
himself has been quick to say all along.

second, at some point in time, decisions
have to be made that are not about "errors"
per se, but which flow from the philosophy
with which you are attacking the task at hand.
for instance, when a specific chapter-header
fails to match its entry in the table of contents,
do you change one of them to _force_ a match?
well, it depends on what you want to be doing...
decisions like these become far more important
than "correcting errors" very early in the process.

third, and the main reason i rolled my eyes and
decided to give no response is because i have
already demonstrated -- countless times here,
right on this listserve -- how to go about the
process of creating an "error-free" digitization:
by comparing two independent digitizations and
resolving the differences between them.  bingo.

so i've been through the routine time after time,
and now michael is proposing we do the routine?
what a strange and curious request.  i didn't get it.

so rather than send back a reply saying "wake up!",
i just opted not to say anything.

at any rate, the reason i bring all of this up now is
because here's a situation where they actually have
two separate digitizations, and they did not even
do a simple comparison of one against the other?

that's stupid, to a degree that is embarrassing...

and -- this is most telling of all -- not surprising.

-bowerbird