There's an old saying that states "A man with a clock knows exactly
what time it is.  A man with _two_ clocks is never quite sure..."

While we might need to copy 3 (or more) rolls to be sure that we
have one good reconstructible master scan of that roll, if we scan
the only copy we have, then we have preserved at least that one copy.
We may never know if there are others around with the same number, etc.
or if they are the same.

Case in point: QRS 995 "Dardanella" (released 2/20, P/B Ted Baxter and
Max Kortlander), in its original release, is very different from the
roll released much later which carries the same information and roll
number.  I only know this because I lucked into an older copy in a
collection: the older version is MUCH better!

So how will we know if the one roll we have scanned is the "best"?
Frankly, we won't, because opinions will differ regarding what
constitutes "best"!  If we try to preserve every known copy of every
known roll by every known maker, that would be truly impractical: we're
building an insurmountable obstacle in front of us!  If we try to get
good scans of every roll we can, we'll be doing the best preservation
we can.

Regards,
Bob Loesch

 [ The man with one clock is blissfully ignorant, whereas the man with
 [ three clocks also knows how precise the indicated time is, and that's
 [ my point regarding copying music rolls.  Rolls punched from the same
 [ master roll frequently display differences due to perforator machine
 [ malfunctions.  The only way to determine how faithful is the
 [ transcription, and hence the data offered for archival, is to compare
 [ three or more production rolls.  -- Robbie