[AV Media Matters] Saving the folk heritage

[smolians@erols.com]

This was sent the listserve of various folk-related archives with an interest
in preservation of sound recordings. Their address is
sound-L@listserv.american.edu.

THOUGHTS ON PRESERVING THE AUDIO FOLK HERITAGE
Steve Smolian
1/19/01
The recent "Folk Music in Crisis" conference at LC and a couple of
panels in which I participated at the National Archives prompted the
following thoughts, some of which I expressed at the time and some
of which are a result of subsequent mulling.

Millions of hours of desirable sound on passe, inaccessible media
sitting in public archives and private closets worldwide make it
imperative that the recordings hospital to which the audio ambulance
takes them can restore them properly and give them new life at the
lowest reasonable cost so more can be saved.

This is particularly so for the huge number of hours of folk
materials in public and private hands. There is a serious effort
underway to save as much as possible of the audio folk heritage. The
Library of Congress, Smithsonian institution and the National
Endowments are about to fund many such efforts provided certain
criteria are met. Identifying those areas requiring criteria and
establishing them is being done at a series of meetings and
discussions this year in which these organizations are
participating. The objective is to establish standards and practices
concerning sound quality and longevity for this type of material to
avoid wasting funds.

There is a sharp division in best recoverable sound quality among old
material stored on obsolete audio formats. Cylinders, wires, dictation belts,
have a fidelity range restricted to no better than poor AM. Discs fall into the
same quality range but can extend to really good AM and, in some cases, FM. Tape
covers the entire range from poor AM (cassettes without Dolby) to record master
quality.

Though a medium may be capable of the highest fidelity, in most
cases, the application for which they are used in the folk context
gives mid-fi results at best. Most interview situations are set up
to get a living stenographic record, adding a specific voice and
inflection to the what may appear bland on the printed page. Field
recordings, which are made under less than the controlled conditions
of an isolated, dedicated studio, will be limited by the properties
of a more casual recording environment. The collector’s
objective is to capture the sound with the greatest quality he can,
taking that environment into account.

Some folk recordings are made under excellent acoustical
circumstances. Perhaps those should be identified and treated at a
higher engineering level during the preservation process. I expect
they make up a very small portion of the enormous backlog, most of
which are aged, fragile, limited fidelity sound carriers awaiting
sonic migration to a more stable medium. Though studio restoration
time remains the same, the somewhat fragile $ 2.00 cassette has been
superseded by the sturdy $ 1.00 CD at a 50% reduction in material
cost. Preservation-prescribed reel to reel tape now runs just over $
20.00. Of course, care must be taken to use CDs with reinforced
surfaces rather than the short-lived silver disc so enticingly
available in office supply stores.

The present CD recording standard of 44,100 samples per second with
a 16 bit word length is a vast improvement over the highest possible
quality from all analog audio carriers.

I venture to say that over 90% of the audio awaiting preservation
and subject-relevant to this discussion is of fidelity no greater
than good but not great AM radio. The frequency range of the present
CD is three to four times that of the material it is to preserve.

The recent group of test result publications concerning CD
lifetimes, two in the Joint Technical Committee Proceedings for the
2000 conference, the other from Australia, do not seem to segregate
those discs with a hardened surface and with the white, printable
one from those with ordinary thin lacquer coverings. It does not
seem to have been made clear (possibly I misread the results) if
these two CD design changes, directly addressing longevity, were
included at all or, if so, were reported upon as separate products.
The figure of an implied or stated five to ten year lifespan is
being used by some to call into question the viability of the CD as
an archivally suitable medium. The test work published a year or so
ago by Eastman Kodak, claiming a c. 200-year lifespan for the coated
disc, is at sharp variance with other test results. Test processes
and conclusions should be analyzed to find where they differ,
determine which give more accurate data, and why. The effect of this
ambiguity is presently crippling adoption of the CD by
standard-setting organizations as a replacement medium for
reel-to-reel audiotape

Higher sampling rates give greater fidelity, but the input must be
able to take advantage of that increase for it to make economic
sense, unless that rate is built in to the commonly used, affordable
system. I know of no dispassionate tests having been run on the
audible differences among the various sampling rates and word
lengths, and certainly none as relates to sound sources that have
been less than perfectly recorded. So forgive my skepticism.
Deferring archival transfer activities is unwarranted while
anticipating new media systems whose improvements are unrelated to
this specific preservation mission. Standards for many of these new,
highly publicized products, do not yet exist and some are yet to
come out of the test stage or are offered only by a small percentage
of the groups addressing that overall market. Archival treasures
will continue to die while we await yet another audio Messiah.

There are those who claim there is sub- and supersonic information
not being captured by transfers to the present digital system and
they may be right. But losing this should be weighed against the
potential to save so very much by using the CD. The increased
quality of "still better" recording systems comes at a great
financial cost. The devices are far more expensive than
today’s CD recorders as is their media. I sense a feeling of
betrayal by the audio community from those who had hoped that the CD
would give them a cheap preservation medium accepted by the archival
watchdogs within their own institutions and fall within the
guidelines of the national funding organizations. All this talk
about "new and better" means "out-of-reach expensive" to all but the
wealthiest institutions and is in danger of creating a fissure
between those few and the rest of the sound archiving world. This
would be quite ironic, considering the fresh preservation emphasis
being placed on folk materials.

Though the audio restoration process has not become simplified,
making copies of the finished results is easier as the appropriate
software and CD recording devices verges on institutional
universality. Once the CD bearing the audio transfer has arrived,
back-up copies, indexing, complete labeling and other such tasks can
be accomplished within the archive, usually at a cost lower than
having this work done in a professional studio.

Sound that interferes with the clarity of the program material is
presently dealt with by a variety of digital audio workstation
products. These programs introduce a change of audio quality- a
dulling of the sound and loss of room "presence". This became
abundantly clear in my own studio when I conducted a series of
informal tests, comparing befores and afters. First, I ran the sound
I had just recorded from an old 78 through one of the better
workstation systems with the results I just mentioned, then went
back to the raw file and declicked using the pointer, scissors icon
and mouse and found no such loss. It thus stands to reason that the
problem is in the widely used algorithms in many work station
programs. I don’t know if this is true for the present, quite
expensive Cedar system as well, as I have not had access to one. It
should be uniform practice, adapted by the memberships of all
organizations involved with archival audio preservation, that the
original be transferred without using automated cleanup programs
and, if desired, a service, subjective, non-preservation intended
copy be made later which may use them. This does not rule out
editing fragments from damaged, discontinuous recordings into a
whole performance. I’ve heard it stated that the greater
sampling rates allows smaller wave forms and makes it easier to
declick automatically though this has yet to be demonstrated. And
maybe we just need to rethink and reengineer how the present systems
work.

I work manually at the one or two bit level which gives a clear
enough look at the problems for me to be sure the issues are not
those of click length but the ability of the programs to recognize
and inaudibly remove other unintended noises.

In summary, I feel the use of expensively available technology
capable of dealing with the greatest extremes in fidelity are
oversuited to the mass of material awaiting preservation in most
folk-related collections. Equipping archives with a supermedium uses
so great a portion of available funds that much less will be saved,
albeit in marginally better sound. The solution is to get as much
audio in indifferent sound on deteriorating carriers migrated to a
modern format or system as quickly as possible.

Resolving the confusion concerning test results on CD longevity
should be a major priority and include analysis of conflicting
published data as well as further testing. The issue of different
high-resolution sampling techniques being discernible by real humans
when applied to older recordings must also be addressed as well as
that of signal alteration by sound clean-up systems. Professional
organizations involved in audio preservation should be encouraged to
adopt a set of standard practices that keep the audio intact.

A substantial effort toward preservation of traditional music and
oral histories on sound recordings is in the process of being
funded. The costs absorbed in saving them are dwarfed by the
mountain of old media awaiting transfer. It is imperative that the
funds be used to save as many as possible of these as diamonds in
the rough, provided what is passed on allows future polishing.

From: Jim Lindner <jim@vidipax.com>
Date: Thu, 18 Jan 2001 17:11:30 -0800
To: AV-Media-Matters@topica.com
Subject: [AV Media Matters] Quantegy
Message-ID: <0.10003996.939229243-951758591-979866653@topica.com>
MIME-Version: 1.0
Content-Type: text/plain

Their web site does not seem to be working - anyone with any ideas why?

Jim Lindner - President
VidiPax - The Magnetic Media and Information Migration Full Services
Company
Telephone 212-563-1999
www.vidipax.com
Moderator of A/V Media Matters@topica.com