Matthew Caulfield's experience is appreciated and valuable from several
points of view.  First, he mentions that he has taken these notions for
granted.  He probably assumed anyone serious enough to put hundreds of
hours into a project would become knowledgeable about the field also.

This is sadly not necessarily so.  If it were, Apple Computer would have
never butchered the language and technology of printing, composition and
typography, planting a technical cancer which has demolished the craft
in one decade.

We needn't think we might destroy the field of library science from
the piano-roll end of the subject, but we owe it to ourselves that
the work of cataloging is made more valuable by being formatted in a
transportable way.

Welte, Ampico and Duo-Art may have actually tried to make their systems
incompatible, but we would be wise to do the opposite with our cataloging.
Let us use the method and language of library science, incorporating the
best practices in data processing and avoiding blunders like the wildly
different implementations called the Dewey Decimal System.

Matthew said:

> I'd call the work a catalog, if it's aim was limited to listing
> and indexing in an organized fashion a particular person's or
> organization's holdings.

This goes for any listing of items, a manufacturer's output or an owner's
library. This is a data handling task, a listing of facts as they stand.

> I think of a rollography as a systematic listing of all known rolls
> intended to form an authoritative guide to the subject.  Which means
> that source errors would be noted and corrected.

A piano roll parallel to a discography, this seems to describe a critical
review or discussion of a subset of the "master" catalog. This could be
literary exposition, perhaps discussion, opinion or conjecture.  A listing
and critical analysis of the Gerswin rolls or of the spark-gap recordings
from Dr. Hickman's machines, or all the rolls with "jass" or "Apollo" on
the label.

If we were cataloging books, sheet music or manuscripts, we would have a
number of equally relevant fields to fill. Most library cataloging methods
stop short of the long list of fields we consider relevant on a piano
roll. Therefore it is important that we decide what fields are needed and
how they will be handled.

A complete master catalog of every roll variant created, even the one I
did with the X-Acto knife, would be a necessary reference to have Robbie
get his wish of putting in the "62583" as a catalog number and choosing
which "62583" from which to accept the first dozen fields.  What a de-
lightful concept! We are really not that far from being able to create
most of this.

In his own catalog, Robbie has most of the fields Matthew mentions.
A database with flexibility to include multiple entries in each field is
very valuable for this work, since there are likely so many multiple
authors, players or tunes on the same roll. Some less obvious character-
istics would be very valuable too: The same note record on three different
 systems and several labels would be a valuable thing to correlate.
Different note records or coding issued under the same catalog number
would be another valuable abberation to identify. Much of this can be put
in the note field and found by a search program, while the basic fields
could be indexed for rapid retrieval. I have added some stuff ("+" and
"{}".).

> * manufacturer/label/number series {recut info here?}
>   roll or item number¶
+   system (88, welte green, D-A, 65... Some mfgrs made many types.)
>   title
>   composer(s)
>   performer(s)
>   type of composition (waltz, overture, march, etc.),
>   dating (of roll certainly, possibly also of underlying composition)
>   recording location information¶
+   all items below are for an inventory catalog, right?
>   location in collection (for inventory catalog)
>   known owners (for a 'locating' or 'existence' list
>   physical condition/characteristics (for catalog) {"recut" here?}

Be careful about condition -- it is time-variable.

> Anybody in his right mind who is undertaking a large catalog or rollo-
> graphy had better give serious consideration to the method he chooses
> to support his data base.

 { Mount Soap-box } [ Wave Banner! ]

Cataloging has occurred and will continue to occur willy-nilly.  The only
defense to anarchy is an Available, Bullet-proof, Convenient method which
will make it absurd to write anything but a shopping list with any other
method.  This has been the principle behind virtually all widely adopted
standards.  Let us work to define a just such a method.

 { Dismount Soap-box }

Karl Petersen¶
Meridian, Idaho