From: Patent Tactics, George Brock-Nannestad
Karl Miller, Don Andes, and others have discussed this, not the least in view
of the work of archeologists.
Unless somehow, miraculously, every existing sound recording gets digitized
and maintained forever by migration, there will be a loss over time.
Alternatively, everything may be transferred to analog metal masters and
stored with few environmental requirements, along with information of how to
extract the sound. Shellac-based slate-dust records (78s to you) may serve a
similar purpose, as repeatedly and correctly pointed out by Steven Barr.
So, unless one or both are done we shall have a loss in the absolute sense,
but not necessarily perceived as such - if an item will never be demanded,
nobody will ever know that it is missing.
Before we get bogged down by the enormity of the task, let us look at what a
sound recording can bring us at some indefinite time in the future. I am
bringing very basic considerations, but they are useful for getting a grip on
But before planning on a strategy for maintaining sound recordings we must
have an idea of what we expect to get "in the other end", that is at the time
we access the stored sound recording.
If we have, say, 3 minutes of sound, what may that represent?
1) it may be a "blind" recording, just 3 minutes of sound surrounding the
1a) it may like 1) but with a directional microphone
2) it may be a recording of a complete utterance lasting 2½ minutes, with
pure soundscape on either side
3) it may be a recording of a longer utterance, i.e. an excerpt of 3 minutes
4) it may be an edited-down version of a longer utterance to take up
precisely 3 minutes
5) it may be a construct: recorded faster for slow reproduction in order to
obtain Mickey Mouse (TradeMark) voices
6) more generally: an edited sound that never existed in real life
When we reproduce our recording, we may have different purposes in mind:
a) to get an airborne sound that is identical to that present when it was
b) to get an airborne sound that is identical to that which would have been
available to contemporary reproducers of the recording
c) to get an airborne sound that is identical to that which reproduction on
any later equipment may provide
d) to get an airborne sound that is identical to that which transmission via
any kind of network, such as broadcast (medium wave, FM, telephone wire,
etc., early A/D, D/A) would sound like
e) to get a signal that is useful for scientific analysis
Seeing a) - d) we realize that 4) - 6) above may actually be re-recordings
Recording type 6) actually provides the least of general interest, because it
is an edited recording made out of little pieces. It can only tell us
precisely how someone wanted it to sound, but not how something like that
sounds in actual performance. This means that any scientific analysis based
on the time function will be faulty, because the time axis has been cut and
pasted. Hence, we can study taste or ambition as it was at the time of the
original edit - is that enough to maintain its availability?
In a similar line of thinking, we may go through any combination of recording
and reproduction and consider if we think that we want to take responsibility
for the long-term availability of such combinations.
I have on other occasions proposed a coarse division of the field "preserivng
sound", and I have suggested the purposes as "nostalgia", "audio history",
and "scientific analysis". There is absolutely no value judgment involved in
these labels, and they may not be all that are needed. A reasonably complete
discussion may be found in a paper I gave at the 103rd AES Convention in New
York 1997 (I gave three papers during that convention), with a long but
precise title: "Applying the Concept of Operational Conservation Theory to
Problems of Audio Restoration and Archiving Practice", Preprint No. 4612. In
it I attempted to undo some of the damage that had been done to the field by
simplistic models presented in the early 1990s that appealed immensely to
I know that the cost of obtaining this preprint from the AES will put
somebody off, but the general philosophy of Operational Conservation Theory
may be found for free as a pdf-file at
ALL ON ONE LINE !!!
My personal view is that if there is a community that swears by analog
recording and reproduction, they should see to it that there is a solid
manufacturing support for it, i.e. both carriers, components and equipment.
We have a similar situation in the photographic field: the manufacturing
support for film-based emulsions is withdrawing at an alarming rate.
Originally, perforated film for 24x36 mm was basically movie negative film,
but the range available now, due to the demise of B/W moviemaking brings
tears to your eyes. It is the large-scale coating of a strip that requires
manufacturing facilities. Within the foreseeable future you will have wet
collodion plates once more, something that can be made up by anybody by
extremely well-documented, albeit old, processes. Similarly, the only
analogue recording medium that can be prepared in a cottage industry is the
lacquer disc. Wire recordings on nickel would be quite feasible, and
retracing the old principles with modern components is certainly doable. I
shall go the "Whole Earth Catalogue" to find my components.
So, instead of hearing the contiuned whimpering about the demise of analog, I
expect strong voices representing parties willing to pay for continued analog
access to analog recordings - and if needed - continued acces to recording by
analog means. Those who have digitized too early (and it is always too early)
are suffering now and in the near future. Those who have saved decayed audio
by going digital have bought a little time.
Looking forward to further insightful contributions on these matters,