To make an obvious point that I don't think has been made so far: only
one (good) copy of each roll needs to be scanned.  That scan, or rather
better, a reconstructed master made from the scan, can then be shared.

 [ Trouble is, 3 original rolls really should be transcribed so you
 [ can decide if the data of all 3 originals agree well enough that the
 [ transcription may be declared "good".  Think of the worried fellow
 [ who wears 3 wristwatches.  ;-)  -- Robbie

The dog-in-the-manger collectors are a commonplace: it's human nature,
and an electronic gizmo's no match for millennia of selfish genes.
These collectors are also mostly an irrelevance.  Scan somebody else's
rolls and share them instead!

Obviously, it's a problem if they have "only known copy" rolls.  But
I wonder how many of these there really are?  Some, obviously, but
surely not all that many.  Things have a habit of turning up when you
start looking for them, and why should the thrill of the hunt be denied
to roll-scanners?  It's more work, but good fun with it.

Sharing the scans?  Sharp intakes of breath all round at this point.
However, as our moral high ground seems to be that collectors should
share their rolls, it's only fair dealing to share the resulting scans!
I'd like to quote from Richard Stibbons' first posting to the
rollscanners group which states the position rather nicely:

  "... Who does the result belong to?  There are thousands of copies
of that roll in the world.  Do I keep scanning it over and over, each
time with a different owner's name on?  In the end I decided the scan,
although it involves a load of my work,  isn't really mine at all or
even the roll owners.  That would be like making a photocopy of
"Hamlet" then claiming you're the greatest writer who ever lived.

  "My position, therefore, which I believe is the only practical one
in the long term, is that I'll make everything available to everyone.
If people want to take those scans and make new rolls, that's the best
possible outcome.  If this thing works out, I foresee a situation where
those who run perforators won't need to scan rolls.  They'll simply dip
into our free library of high quality scans and take their pick.
Others will find different uses.  [...]

  "There'll be freeloaders. So what's new?  Does it really matter?
The satisfaction comes from knowing that we are conserving a stock of
exquisite music which, without our efforts, will largely disappear over
the next twenty years.  To those who see roll scanning as  commercial
opportunity I say :-  "Be quick - give us a year or two and we'll be
giving your stock away for nothing" "

Idealistic, certainly, but to an extent it's happening already and the
pool of scans will continue to grow.

Julian Dyer