CONCERTED ACTION ON MUSIC INFORMATION IN LIBRARIES

WP 3: Networking and Digitisation

Deliverable 3.1

Analog Documents, Carriers and Formats

Project: PROLIB/HARMONICA 10453
Commission of the European Communities
LIBRARIES PROGRAMME

Version   :  1 
Date        :  May 30th, 1997
Author   :  OEAW-WAD 
Confidentiality  : Public 
Status     :  Final

3.1.1   Introduction
3.1.2   Physical Deterioration of Analog Documents
3.1.3   Printed Documents, Manuscripts etc.:
3.1.4   Conversion and Transfer
3.1.5   Workflow Analysis Necessary
3.1.6   Still and Moving Images:
3.1.7   Sound:
           3.1.7.1 Cylinders
           3.1.7.2 78 rpms
           3.1.7.3 Acetate Discs
           3.1.7.4 LPs (Vinyl Discs)
           3.1.7.5 Magnetic Tapes
           3.1.7.6 Compact Cassettes
           3.1.7.7 DAT Tapes
           3.1.7.8 CD, CD-R
3.1.8   Analog Recording and Reproduction Devices
3.1.9   Analog Formats
3.1.10 Analog Carrier Storage and Preservation Issues
3.1.11 Summary and Further questions
3.1.12 Further Readings (to be continued)
3.1.13 Annex A: Questionnaire

3.1.1 Introduction

The most important prerequisite for the development of large scale music databases and information services is the transfer of analog documents into digital format (digitisation). Libraries archives and collections which require such large scale transfer of information must develop a purpose built infrastructure. In order to provide institutions with information for developing such an infrastructure more easily and straight-forwardly, the Workgroup No. 3 of Harmonica has drawn up the following guidelines.

In order to obtain trends and figures concerning analog analog and digital media being stored in libraries and archives, a questionnaire has been developed which is intended to producean overall view of music information, other related media and document types.  As well as content, the questionnaire also investigates issues of access and availability. The questionnaire considers the issue of copyright by differentiating between published and unpublished documents, between printed documents published prior to 1920 and those published later on. In case of copyright sound documents, a distinction between those made prior to 1947  and those made after has been introduced. Unpublished historic music information and recordings are prime candidates for immediate digitisation, whereas more recently published documents usually require difficult copyright negotiations when put onto network servers. The figures obtained from music libraries and archives in response to this questionnaire should provide a tentative insight into the holdings of these collections and the relationship between the different categories of information they contain.

Note: the current figures listed in the following tables are preliminary. They are based on the extremely small number of questionnaire responses obtained  up to date and will be completed as soon as further responses from libraries are available; therfore no general trend can be extracted. Statistically speaking, data obtained shows high inhomogenity. Values which can be assumed to reflect a range are written „from - to"; those indicating different library categories are given „xxx,yyy".

3.1.2 Physical Deterioration of Analog Documents

The vulnerability and instability of analog document carriers as well as the technical obsolescence of historic reproduction devices impose a certain urgency to the task of copying. While the copying of books, paper and sheet music material reproduces text and textual content without significant loss, analog copies of sound and video suffer inherently from a  decrease of signal-to-noise ratio. Today historic sound recordings may be digitized to a very high standard of reproduction. Modern analog photographic reproduction techniques provide sufficient resolution to meet the very exacting standards imposed by archival criteria but reproductions of this type suffer from  a limited life-span. Their digital counterparts require extensive disc storage capacity in order to obtain almost the same image quality. Due to the fact that network band widths are limited, only browse copies with reduced resolution can currently be transmitted. In librarian and archival practise, one and the same document has to be stored in compressed as well as uncompressed format. 

---

3.1.3 Printed Documents, Manuscripts etc.:

 

Media Type:	unpublished and rare (historic)  previous to 1920	unpublished and rare (historic)  later than 1920	published (or copies of)
a) Books	10%	15%	10-15%
b) Manuscripts	10-20%	20%	1-5%
c) Articles	-	-	0, 8.5%
d) Music Scores	75%, 30%	60-75%	85%, 0
e) Autographs	5-10%	5-10%	-
f) Other	-	microfilm, database	microfilm, database

Because destructive usage has to be reduced to a minimum, historic books, manuscripts, letters, public papers, reports as well as musical scores, etc. are generally not accessible without the assistance of library staff. Special preservation measures are applied to autographs and other primary sources such as restricted use, separate authorization to see originals only being given in exceptional circumstances.  The formats of printed documents and manuscripts, ranging from pocket-size editions to large sheets such as DIN A2/A3, also impose their own conservation and handling problems. Large volumes are invariably difficult to handle and often cannot be copied without mechanical damage. Elaborated methods for photocopyin and scanning of vulnerable books have been developed. Nevertheless, these restrictions in usage do not stop the continued deterioration of documents due to physical and chemical processes accelerated by environmental influences. For documents such as newsprints, journals and other works printed on acidic wood pulp paper - as most publications have been since 1850 - the remaining lifetime has to be estimated as not longer than a couple of decades. Books from the 18th century or before have more durable paper and bindings. Highly sophisticated conservation measures such as de-acidification programs [1]  can only be provided for a limited number of documents. Bulk de-acidification is still in an experimental stage, while page-by-page de-acidification is extremely expensive. Microfilming has been favored as a way of preserving the content of printed documents. Many libraries still lack suitable copies of printed resources their users would be interested in so that large quantities of rare documents are left without any care of adequate handling and storage.

3.1.4 Conversion and Transfer

Today, the process of digital scanning and the process of microfilming  printed documents are both comparably expensive activities. ASCII conversion of a captured text image or re-keying of text costs about 10 times more. It is expected that OCR technology will make advances, and it will become possible to process digital images and convert them to ASCII in  a way that can be routinely used by libraries. Nevertheless, for some time to come, it will be too expensive to convert to full ASCII.  Sound and video transfer as well as compression and enhancement techniques are also being continuously developed for routine use by libraries.

3.1.5 Workflow Analysis Necessary

Due to the fact that many cases no elaborated and standard procedures exist  for storage and access of printed material in all its variety of formats, bindings, packing and different uses, an  analysis of traditional storage and access procedures is therefore surely predicated. Such an analysis should include both bibliographic and  catalogue searches as well as all actions of library staff members necessary to transport documents from stack to user and back again (taking  shelving, handling and circulation concerns into account). Additional staff activity and copyright clearance is needed whenever copying either from paper documents or microfilm rolls is required; this too must be considered. It has become common library practice to copy frequently requested documents in advance, and most libraries have already elaborated sophisticated strategies to minimize workload and to fulfill standard user requirements.  Nevertheless, costs continue to increase. In digital libraries, this workload is dramatically reduced. For document acquisition and cataloguin procedures for incoming documents, a similar standard procedure should be created, taking expenditure of staff and technical infrastructure  into account. 

---

3.1.6 Still and Moving Images [2]

 

Media Type:	unpublished and rare (historic) previous to 1920	unpublished and rare (historic)  later than 1920	published (or copies of)
a) Still Images	100%	98%	90%
b) Moving Images	-	-	0.5%
c) Video	-	-	0, 99%
d) Multimedia [3]	-	-	-
e) CD-ROM	-	1%	1-4%
f) Other	-	1%	-

Cellulose nitrate film is known to be extremely unstable; it decomposes with age and is highly inflammable. It has, however, unfortunately been the medium for quite a number of music and theater film production recordings up to 1950. Due to its age and its flammability at relatively low temperatures, it has - for a long time - been the focus of preservation attention. Its lifetime can be extended to more than 90 years under proper storage conditions [4]. In contrast to this, the cellulose acetate „safety" film that replaced nitrate and even laser discs, has shown occasionally shorter lifespans. For many years, nitrate film was considered discardable after having been copied  onto safety stock.  Archives and studios have since, however, re-thought this policy, because even the best currently available safety-film copies have proven incapable of reproducing nitrate film's subtle visual qualities. The original film has maximum image resolution and sound quality and - if stored accordingly - can remain the best source for copies in any future format for a long time. Except when dangerously deteriorated, nitrate could be retained for re-use as duplication technology improves [5]. Last not least, rapid fading of recent "dye-coupler" colour emulsions that became standard after 1953 has been recognized as a problem for today’s archivists.

Little or no information is available on how to preserve music television film and video products. Since the introduction of television broadcasting, archives have moved rapidly into collecting 16 mm film, videotape and other formats - often rescuing material which would have been thrown away by public or private television stations. As video has become more portable and inexpensive, many music recordings for broadcasters internal documentation and educational outreach have been created. All these materials are now stored in archives, many of them in obsolete formats. The formats range from video disks to 1/2-inch and 3/4-inch cassette; to 1/2-inch, one-inch, two-inch open reel etc.

Video preservation faces similar problems of obsolescence of hardware and instability of magnetic tape (cf. 3.1.7 Sound). The inevitable conclusion of investigations into the preservation of musical information stored on video, still images and film is that video, still and motion pictures of all types have been and are being produced faster and in ever increasing volumes. Consequently, they will deteriorate faster than archives will be able to preserve them.

3.1.7 Sound:

 

Media Type:	unpublished sound  documents	published sound documents previous to 1947	published sound documents  (or copies of) later than 1947
a) Cylinders	20%	3%, 0.1%	-
b) 78 rpms	50%	97%-100%	10%
c) LPs	0, 4%	-	50-75%, 15%
d) Reel to Reel Tapes	5-15%, 70%	-	5%
e) Compact Cassettes	1%, 50%	-	5-15%
f) CDs	2%	-	10-20%, 80%
g) DAT Tapes	0.5%, 25-30%	-	0.8%
h) Sound files, digital	0%, 10%	-	n.a.
i) MIDI files	-	-	n.a.
j) Other	0%, 30%	-	-

3.1.7.1  Cylinders

Among the most vulnerable sound carriers are acetate discs, wax cylinders and wax discs as well as acetate-based magnetic tapes; all of these suffer from progressive decomposition [6]. Their lifetime is limited to a few succeeding years only. They have to be transferred into digital formats as soon as possible. Record companies used wax cylinders as recording carriers from the earliest sound recordings in 1889 to the late twenties [7]. While many of the approximately 300.000 cylinders (world-wide) [8]  have been copied to modern analog media, those remaining uncopied - many of which contain rare and valuable material - are in immanent danger of being lost forever.

3.1.7.2 78 rpms

Coarse groove gramophone discs (78 rpms) carry approximately 10 million audio productions from 1898 until about 1955. They are  physically fairly stable and show no chemical instability. Nevertheless, improper handling may result in mechanical breakage.

3.1.7.3 Acetate Discs

A surprisingly large number of acetate discs [9] about 3 million recordings world-wide - many with unique contents - will  probably be unavoidably lost within the next few years. Acetate discs were used until the late forties and early fifties as an instantaneous playback recording, medium comparable to the reel-to-reel tape which emerged later on. An inherent hydrolytic chemical reaction causes the lacquer coating, which carries the grove, to become brittle and separate from its aluminum core and consequently the information is lost (Fig. 1). Up to date, no effective conservation measures have been found to preserve these recordings. Intact acetate discs are, therefore, considered to be the highest priority candidates for careful copying.

 
Fig. 1: Acetate disc showing lacquer coating flaking from its disc core.
Photograph, Phonogrammarchiv, Vienna..

3.1.7.4 LPs (Vinyl Discs)

The total number of microgroove discs stored in music libraries and sound archives is more than 30 million world-wide. No special instability problems have been reported to date. They have proven to be the most stable of the materials that have been used in the manufacture of sound recordings [10].

3.1.7.5 Magnetic Tapes

One of the most popular post-war recording media was reel-to-reel acetate tape. At that time, longevity was not a prime consideration in the development of tape recording methods and, therefore, severe conservation difficulties inevitably arise. After only a few decades, the base becomes brittle and may simply break as soon as playback is attempted. An extremely unfavourable example of what may happen is given in Fig. 2. Modern magnetic tapes (audio and video) based on polyester film are considered to be relatively stable under favorable environmental conditions. Nevertheless, some problems have been reported concerning the instability of pigment binders. Atmospheric humidity is absorbed by the pigment binder causing the polymer to hydrolyze. Special attention has to be paid when smearing of magnetic oxide particles on tape machine heads is observed during playback. Transfer procedures include the control of electrical distortions due to chemical and physical deterations.

3.1.7.6 Compact Cassettes

Compact cassettes are not used for professional archived copies, this format is predominantly present in the home consumer market. Consequently, there is a hugh amount of sound recordings stored in that format. Preventive preservation and security procedures are comparable to those used for modern magnetic reel-to-reel tapes. Due to smaller recording track width and speed, the overall signal-to-noise ratio of compact cassette recordings may be reduced. As consumer sound recordings are generally not made under controlled recording conditions as expected from carefully serviced studio equipment, misalignment of recording heads vow and flutter of the tape transport frequently cause further sound quality reduction.


Fig. 2: Severe case of pigment shedding of a modern tape suffering from pigment binder hydrolysis after playback has been attempted. Photograph, Phonogrammarchiv, Vienna.

3.1.7.7 DAT Tapes

DAT tapes should not to be considered as a secure long-term storage medium, and are mainly used as an interim mastering medium. Digital streamer (DAT) tapes are similarly used for short-term backup purposes of computer hard disks. There is no principal carrier difference between (audio-) DAT and digital data streamer tapes except that a higher quality level for streamer tapes can be assumed. Digital data streamer tapes are written in different formats and usually allow individual file access. Digital streamer DAT´s have now become the most popular backup medium for small size server capacities between 8 and 32 Gbyte. They are supported by standard system backup software in commonly used systems like WINDOWS 95/NT, UNIX etc.

3.1.7.8   CD, CD-R

Audio CDs as well as recordable CDs (CD-R) suffered from delamination and reflective layer corrosion during their introductory phase. Today, rough handling still generates problems. The longevity of CD-R is still a matter for discussion and few of the manufacturers of blanks publish sufficiently detailed figures. One exception is Kodak, who claim a minimum life expectancy of a hundred years for their CD-Rs, and back this up with detailed maths. In comparison to magnetic tape, CD products probably provide longer life expectancy and make it easier to generate further redundant copies (master copies and derivates). A master copy should be stored properly and only handled to produce derivative copies. Special attention has to be given to the life expectancy of unwritten blanks. There are cases where lifetime has been reported to be limited to approximately 2 years, depending on manufacturer and product. Last not least, the use of high-speed CD-blanks in low-speed CD-writers may cause problems.

3.1.8 Analog Recording and Reproduction Devices

Whenever a machine is necessary to view, to read or to listen to analog documents, users and libraries depend on the functionality of appropriate reproduction devices. Several generations of tape recording machines with different tape speeds and track formats must be repaired and serviced in order to work properly when needed. Audio and video recorders must be maintained in excellent condition in order to reproduce high quality recordings and to prevent damage to tapes during playback [11].  The mechanical and electronical accessories are equally obsolete as the devices themselves. Libraries and archives have little or no influence on the developments of new industrial AV-technologies upon which they depend. Since the mid-eightees, the industry has also changed its policy. Wheras in „the good old days" of the audio industry, professional studio equipment development was forerunner of consumer products, nowadays the professional version of sound studio equipment is frequently developed on the basis of consumer AV-devices and formats. New reproduction devices frequently do not guarantee backward compatibility neither in electronics, recording formats nor in data carrier mechanics. The reconstruction and maintainance of obsolete reproduction devices and -conditions is extremly costly and inffordable for many libraries and archives [12]. Technically expensive preservation issues may, therefore,  be transferred to institutions and facilities specifically dedicated to preservation techniques and to business-oriented companies. But, library staff, who understandably feel responsible for their collections are reluctant to outsource their digitisation or preservation tasks to outside organizations, sometimes develop even  psychological barriers which delay prohibit re-recording and preservation activities.



Fig. 3: Custom build cylinder machine equipped with modern pickup and
amplifier electronics. Phonogrammarchiv, Vienna [13].

3.1.9 Analog Formats

Format progression (obsolence) is primarily an issue for audio-visual media. Classic text and image documents on paper as well as photopraphic media have changed formats without substantial consequences for the retrieval of contents. On the contrary, AV-documents in obsolete formats are in danger of becoming unplayable, with an  irretrievable loss of stored contents, although the lifetime of the information carriers has been extended. For example, analog sound and video formats on reel-to-reel tape or cassette range from 4 mm, through 8 mm, 1/2-inch and 3/4-inch cassette; to 1/2-inch, one-inch, two-inch open reel etc.

3.1.10 Analog Carrier Storage and Preservation Issues

Low-temperature, low-humidity storage, low shortwave light and low pollution levels are key conditions for a balanced preservation strategy. Costs for the construction of storage facilities and their operation are high but they are justifiable in view of the overall costs of preservation. State-of-the-art storage facilities now aim at maintaining film which has reached a critical stage of deterioration at temperatures ranging from 2 to 8°C and at a relative humidity between 35 and 45%. Even small decreases in temperature and humidity have been known to allow substantial extensions to the life of film stock. Magnetic tape material is less vulnerable to temperature fluctuations than to increases in storage humidity. Reasonable (affordable) access storage values are 17 to 20°C at a relative humidity below 45%. Due to the fact that improving storage environments is an expensive but  less visible and less dramatic task than restoration and copying programs, it requires longer-range planning.

Numerous research laboratories and organizations have moved into the field of conservation and preservation work. A comprehensive directory of information sources on scientific research related to preservation can be obtained from

http://palimpsest.stanford.edu/cpa/reports/child.html.

3.1.11 Summary and Further questions:

In practical terms, the advice for libraries and archives can be summarised as follows:
• The most expensive parts of preservation and transfer activities are (1) Selecting the documents to preserved, using criteria developed from risk evaluation as well as access and user requirement studies (2) The physical task of turning the pages of the selected book for item-by-item chemical treatment, filming, or digitizing (3) Preparing the appropriate reproduction equipment and selecting the tapes, discs or cylinders for high quality controlled transfer etc. To produce a microfilm image, digital scan or audio file, the major cost is expended in the time required to gain access to each item so that each book, each film, each tape etc. is handled only once.
• To assure archival copying quality, document format specific plans are needed to recommend and review laboratory preservation work, establish technical guidelines and  to develop generally accepted procedures for large preservation and transfer programs. These must take thee new scientific understanding , of document deterioration and transfer technologyy into account. Several guides concerning the care and handling of magnetic tape, books and images which are already available.
• Libraries and archives must recognize the importance of saving the original discs, original film and images, even after copying, unless they have deteriorated beyond any use. Saving the original documents remains a basic principle.
• Cost plans for analog and digital document storage and access should be elaborated in order to provide realistic basic assumptions for a libraries business model.
• Libraries and archives still work in isolation, frequently duplicating one another's efforts. There is, unfortunately, no European cooperation - neither in the field of printed document preservation nor in film or sound carrier preservation; most initiatives are restricted to national level. Different national preservation programs should be coordinated, ideas should be shared and an European strategy of cooperation should be developed. Representatives across the library and archive community, from nonprofit and public libraries and archives, as well as representatives from music information publishers, the motion picture and recording industries, research laboratories and universities etc. should all participate.
• In times of shrinking public funding, music information preservation and initiatives to make music documents more accessible to a wider public require increased EEC support as well as private sector partnership. Combined public-private funds could establish the foundations for new relationships between archives, the entertainment industry, the educational community and the general public.

Some questions to be answered:

• Should traditional procedures of archival storage and access be further applied? For how long?
• What are the overall costs of traditional library and archive usage?
• Are there hidden benefits to digitisation? Now, from the beginning? In five years from now?
• Would introducing new digital technologies in libraries and archives requires additional library staff and new training?
• Shall enough technical expertise be available in future?
• Do we have the educational potential for that? Do we need new scholarships?
• Are necessary changes of storage and archiving philosophy are on the way.

Acknowledgement:
the authors express their thanks to Ben Dhaliwal and Herta Spielmann for the English Language editing of this document.

3.1.12 References:

[1] See project MUSE, Commission on Preservation and Access: http://palimpsest.standford.edu/cpa/newsletter/cpanl82.html

[2] Reilly, James M. and Franciska S. Frey (1966): Recommendations for the Evaluation of Digital Images Produced from Photographic, Microphotographic, and Various Paper Formats. Reports to the Library of Congress National Digital Library Project contract # 96CLCSP7582; http://lcweb2.loc.gov/ammem/Ipireprt.pdf.

[3] A distinction between Multimedia Presentation Files (c.f. Adobe Premiere, Toolbook, Power Point) and CD-ROM-like productions has been encountered.

[4] FIAF, Brussels (1994); Philosophy of AV Archiving.

[5] CPA; Recommendations of the Librarian of Congress in consultation with the National Film Preservation Board (Aug. 1994); co-ordinated by Anette Melville and Scott Simmon:  http://lcweb.loc.gov/film/plan.html

[6] Copeland, Peter (1995): Project Digitise. Final Report to the Digital and Network Services Steering Committee. British Library, National Sound Archive; London.

[7] Schueller, Dietrich (1996): Safeguarding Audio and Video Recordings in the Long Term. In: Stephen Foster (Ed.), Proceedings of the 1st National Memory of the World Conference, Oslo, 3-5 June 1996. UNESCO (Paris) 1996, 131-140.

[8] Gibson, G.D. (1994): Audio, Film and Video Survey. A report on an international survey of 500 audio, motion picture film and video archives. Library of Congress, Washington DC.

[9] (instantaneous discs)

[10] see Commission on Preservation and Access (CPA): The Care and Handling of Recorded Sound Materials; http://palimpsest.stanford.edu/cpa/reports/sound.html

[11] see Commission on Preservation and Access (CPA): Magnetic Tape Storage and Handling. A Guide for Libraries and Archives. ISBN 1-887334-40-8.

[12] Custom built cylinder machines equipped with modern pickups have been developed by the Phonogrammarchiv of the Austrian Academy of Sciences, Vienna and the National Sound Archive, London.

[13] Lechleitner, Franz (1997): An Audio Pre-Amplifier Designed for the Transfer of Historic Sound Recordings. In: G. Boston (Ed.); Preserving the Audio Visual Heritage. Proceedings of the 4th Joint Technical Symposium, London, 1995.

Further Readings (to be continued):

AES (Audio Engineering Society), AES22-xxxx, Draft AES Recommended Practice for Audio Preservation and Restoration - Storage of Polyester-base Magnetic Tape, 1996.

CALAS, M-F. et J-M.FONTAINE (1996): La Conservation des documents sonores. CNRS Editions, Paris, .

Federation Internationale des Archives du Film (FIAF) (Ed), Preservation and Restoration of Moving Images and Sound, Brussels 1986.

International Association of Sound Archives (IASA) - Technical Committee (1997): The Safeguarding of the Audio Heritage - Ethics, Principles and Preservation Strategy.

ISO 3897 Photography - Processed photographic plates - Storage practices
ISO 5466 Photography - Processed safety photographic films - Storage practices.
ISO 6051 Photography - Processed reflection prints - Storage practices .

Music Internet Resources: http://www.lib.berkeley.edu/MUSI/internet.html

SCHUELLER, D. (1993):  Behandlung, Lagerung und Konservierung von Audio- und Videotraegern. In: Das audiovisuelle Archiv 31/32, 1992  21-62.

SCHUELLER, D.(1997): Zur Problematik des Transfers analoger Audiobestaende in die digitale Domaene. In: 19. Tonmeistertagung Karlsruhe 1996, Bericht. Muenchen , 732-733

VAN BOGART, J. (1995): Magnetic Tape Storage and Handling. A Guide for Libraries and Archives. Comission on Preservation and Access, Washington DC.

WILHELM, H, and C. BROWER (1993): The Permanence and Care of Color Photographs: Traditional and Digital Color Prints, Color Negatives, Slides and Motion Picture, Grinnell, Iowa.