It's interesting that a perception has built up that somehow there
is a feeling of antipathy to scanning.  Other than a tiny number of
well-connected highly vocal individuals with their own personal agendas
I don't think this is the case, or ever has been.  Indeed, from those
who express an interest, the response seems unequivocally positive.
I think everyone in this hobby can see some potential value to themselves
in preserving the music and the rolls.  Disbelief or misunderstanding,
sure, but little active opposition.  And frankly, who cares if there
is?  It's like trying to stop the tide.

I am lucky enough to host a roll-scanner designed and built by that
ingenious duo, Spencer Chase and Richard Stibbons, as further fiddled
and tweaked here, and with some spectacular roll-processing software
by Warren Trachtman.  My personal interest is the exact replication of
music rolls -- read my posts amongst others for the last decade in MMD
and you'll see the thread unfolding from pipe dream to project to
reality.

Armed with the 88-note Themodist / Duo-Art perforator built by the late
Steve Cox, after a lot of fiddling (mostly in software and the sourcing
of materials) the setup can roll-scan, process into reconstructed roll
master and perforate all in a couple of hours.  It means that it's
practical to perforate exact replicas of any roll that will sell four
copies.  Those interested may care to have a look at my (very plain)
web site, www.pianorolls.co.uk, and ask more off-list.  Remember that
it's a one-man hobby operation with tiny output!

There are some residual issues in scanning: it can be hard to create
an accurate reconstructed master of some rolls, for instance.  With
some time and a little musical savvy a decent result can normally be
obtained, and things will get better as software evolves.  Mind you,
some rolls never had proper masters, so you do need to know when to
give up on all this finesse.

Far from collectors being chary of seeing their guarded rarities being
copied out to the whole world, on numerous occasions I have actually
been told "I've got these really rare rolls that ought to be copied."
Tonight was spent scanning a number of hard-to-find movements of
Beethoven Sonatas, so that three collections will now have the complete
Sonata, rather than each being missing a roll.

And surely this is what the makers wanted, to be able to buy the rolls
you want.  The idea that they would be happy for someone (say) to be
without a movement of some favourite work "to keep it exclusive" is
perfectly ludicrous.  Not only does the project revive the physical
roll, but it revives the ready access to great music that the
instruments were always intended to supply.  e-Rolls achieve the same
thing, and to a lesser degree so do simulations for other instruments.

To finish on a practical level here are some examples of what can be
achieved.  I have started seeking out the British Duo-Art series which
are scarce elsewhere.  Indeed, many are scare here as well: only one
copy has surfaced in collections we know of 080, "Entr'acte from the
Snow Man," played by the composer, Erik Korngold.  Some of this
material is rather interesting, and some definitely is not (to my ears
anyway).

Another case is where, thanks to some judicious pushing in the right
direction by MMDers, a Cortot roll which for some unknown reason was
only issued in the UK has been located to share it with others; Charles
Davis Smith rejected it as a listing error and left out of his catalogue.
It's 3099, Hungarian Rhapsody No 2, with Cortot's own cadenza.  Not
exactly rare, but certainly obscure.

Julian Dyer