Re: [ARSCLIST] Registry of Digital Masters

[Karl Miller <lyaa071@xxxxxxxxxxxxxxxxx>]

On Thu, 14 Sep 2006, Marcos Sueiro Bal wrote:

> Hard as it may be, choices must be made on what to reformat, or at least on
> what to reformat first; and I think that knowing if an item has already been
> properly digitised could help in setting up priorities.

I agree with you.

> I am not sure that a separate registry is needed. Perhaps all that is needed
> is good metadata in cataloguing records describing when something has been
> digitised properly: that way there is only one place to look. The record
> should also include what you mentioned: What type of file, what kind of
> access, and whether there are Spanish announcements on top of the music
> <grin>.
>
> I agree about your point about multiple copies being one of the best
> preservation methods. But again, I think the info could help _prioritise_
> reformatting.

I am not sure what sort of registry is needed, but I do believe that cost
prohibitive nature of preparing a MARC record is a major factor that
inhibits the use of OCLC as such an information resource.

Consider the normal process of digitization as done by libraries. The
decision to digitize is often based upon...what can I most easily get a
grant to do...how many requests have I had for this item or that...

These criteria have nothing to do with the fundamentals that I believe
should be considered, namely things like the chemical stability of an
item, real and potential historical value (a sort of expertise rarely
found, or seen as important to acquire by libraries), availability of
analog equipment, etc. Uniqueness is of course, paramount, but I believe,
aside from the obvious, such determination can be problematic.

Libraries tend to not seem to see fit to combine subject expertise with
digitization skills...for that matter, libraries these days rarely seem to
place much, if any value on subject skills...So, if you had someone digitizing that
Copland Symphony performance...would they know what side follows what
side, if the discs were pitched properly...difficult to believe? Some years ago one
of our resident library administrator/idiots told me to use technical staff to make
preservation copies of some tapes. I tried to explain to the idiots the
folly of it all...we ended up having wonderful copies of things played
back at the wrong speed...with tape squeal thought as being part of the
music...etc. 1/4 track tapes being played on 1/2 track machines, etc.

Ok, so you outsource the work. How do you know the person doing the
transfer knows enough to tell if that announcement over the end of the
Copland Symphony is just a "normal" way of doing things...even part of the
music. So unless you have some informed staff who can monitor ALL of the
work, how would you know...knowledge required to compile accurate
metadata...a level of depth to the metadata that could address the
"uniqueness" of the recording.

Then, would the person making the dub know if you have a line check, an
aircheck...or second generation copy, etc.

This level of information will not become part of the metadata. Libraries
don't understand the need for it, do not value the expertise to recognize
the need for it, and, quite frankly, don't have the financial resources (I
would argue, due in part to their continued use of inefficient
methodologies for their basic operations...inefficient methodologies which
have resulted in their limited production of access to
information...production which has been overshadowed by the more efficient
methodologies of the commercial world) to do the job even if they had the
inclination to do it right.

> I think the blurb about libraries being committed to preserving their
> artifacts deals specifically with their digital copies. What you do not want
> to happen is the following: "I was going to digitise my record, but Library
> B already did". Then, when you try to get the file later, they say "Oh
> yeah... our server crashed (or whatever) so that file is lost".

Libraries will often "outsource" the management of their preservation
files. Libraries don't even have staff who are fully conversant in the
technology to write contracts to insure the perpetuation of the
information they create.

In short, I agree with you on most things I guess. Would it be better for
us to have but that one copy of the Copland with that announcement versus
the time lost in making a copy of the other recording...well, at least we
might be able to save something else. Yet, as it is now, you couldn't hear
it unless you were at the library that made the copy...and, in most
instances, Library of Congress seeming to be an exception, you couldn't
count on that digital file being there in a format that you could use
effectively, transferred at the appropriate speed...forget about proper
EQ, noise reduction so you can hear it, etc. or even find a listing of it
since the library continues wasting its limited resources trying to deal
with the digital equivalent of a card catalog.

By the way, that Copland Third I had was from a private collection. There
is one OCLC record for that performance. If you look it up, it also
mentions that the performance contains an earlier version of the 4th
movement. That disc and bib record exists because I got a copy from a
collector, made a dub and gave one of our music catalogers (all of whom
are excellent) what I thought to be important information. I knew that was
an earlier version of the 4th movement because I had the subject
expertise. I got the side joins right because I had the score and could
read the music.

When that work was done we had a director who understood and valued such
expertise...alas that is no longer the case. But our new director is not
alone as there are plenty of like minds out there in library schools and
ARL.

In short, I sincerely doubt that libraries, again, with the possible
exception of a few like Library of Congress, will ever provide the sort of
information that is needed to identify potential duplication, or even
provide responsible custodianship of what they hold, let alone, as it
seems these days, even be willing to accept collections that require
intensive reformatting/restoration.

Sure, avoiding duplication is great, but from my perspective, it isn't a
simple matter, or one likely to ever be addressed in a meaningful way by
libraries.

Karl