"In April of 1999, the term 'MP3' surpassed 'sex' as the most-searched-on term at some the Internet's top search engines - a phenomenal achievement for a complicated digital music encoding algorithm devised over the course of a decade by a few scientists and audiophiles in an obscure German laboratory." (Hacker 2000, 1)

German research company Fraunhofer Gesellschaft developed "MPEG-1, Layer III" or MP3 in the late 1980s. MP3 is an internationally recognized codec (codec is an abridged form of "compressor/decompressor." A codec is any encoding algorithm that compresses and decompresses digital information.) that allows users to easily store and transmit near compact disc (CD) quality audio in a digital environment. Basically, the MP3 encoding program takes a standard uncompressed audio file and compresses it to roughly one-tenth of its original size. Therefore a 4-minute song that would normally take up around 40 MB of storage in its uncompressed state would take up only 4 MB as an MP3 file. This is accomplished by utilizing a technique known as "psychoacoustics". Scot Hacker offers the following explanation:

"Uncompressed audio, such as that found on CDs, stores more data than your brain can actually process. For example, if two notes are very similar and very close together, your brain may perceive only one of them. If two sounds are very different but one is much louder than the other, your brain may never perceive the quieter signal. And of course your ears are more sensitive to some frequencies than others. The study of these auditory phenomena is called psychoacoustics..." (Hacker 2000, 22-23)

In the 1950s, when reel-to-reel audiotape hit the consumer market, was the first time in history that people outside the recording industry had the ability to record and distribute music. Next came the rise of cassette tape in the 1970s, which added easy portability to the distribution equation. There was of course one major problem with these technologies: they were analog based. While analog recordings were great at capturing the subtleties of a music performance that might escape the strict "0s and 1s" of the digital realm, it was impossible to make a perfect copy of that recording using analog technology. There was always some degradation of sound quality in an analog copy, and it became increasingly worse with successive copies.

Then came the digital audiotape (DAT). Created in the 1980s, DATs overcame the analog degradation problem by recording the signals digitally, thereby allowing for perfect copies. The major problem for consumers, however, was that the technology was expensive. DAT recording decks retailed for well over $1000, with blank tapes costing between $20-30. While the technology was superior to its analog counterparts, the tremendous price difference kept DATs from surpassing cassette tapes in popularity.

Today there are numerous other music codecs in addition to MP3 (i.e. RealAudio, Liquid Audio, Advanced Audio Codec [AAC]). Not only are MP3s digital, like DATs, they offer another advantage. As Scot Hacker explains:

"Because digital music can be file-based rather than media-based, a single file representing any kind of content can be placed on a Web or FTP [file transfer protocol] server and made available to the entire world at once. The burden of making physical copies, which naturally limited the rampancy of tape-based copies to a large extent, has vanished." (Hacker 2000, 10)

Will the record companies succeed? Since the world of MP3s and their digital music kin are constantly changing, it's hard to tell. Because of this state of uncertainty, librarians should consult the music sections of such technology sites as CNET, ZDNet, and Tech-TV. Other good sites for legal as well as technical issues include the MP3 Newswire and the Music Library Association's Guide to Copyright for Music Librarians.

What does the library community itself have to say about MP3s? While the term "MP3" was never actually used, three essays on the future of music libraries in the March 2000 edition of Notes: Quarterly Journal of the Music Library Association discussed the online digital music phenomenon. Stephen Wright's essay entitled "Technology" stated that:

"Faced with bewildering problems of storage space and licensing, music libraries will inevitably succumb to the temptation to lease collections of digitized audio from commercial enterprises. Undoubtedly this will reduce the stress of absorbing the new audio technology into our collections, but we may also relinquish control over the selection and cataloging of particular works and performances." (Wright 2000, 593)

However, Wright's concerns over selection availability are well grounded. The catalogs of both mp3.com and eMusic.com are aimed mainly towards popular listeners; leaving classical enthusiasts with a limited selection of composers, works and artists to choose from. While it is uncertain what types (and quantity) of music the AlbumDirect site will provide, it is probable that initial offerings will also be geared towards popular music. One can create MP3 files from their own recordings, but there would be no need unless one is mounting these files on a server to allow remote access. In addition, recording or "ripping" tracks from CDs to MP3s is the only truly feasible process as they are both digital in nature. Copying analog recordings (i.e. long-playing records, cassettes) to MP3s would be a much more problematic task.

Tom Moore's essay entitled Sound Recordings put forth three reasons why sound recordings in a physical medium (i.e. compact discs) will remain the format of choice for most libraries when compared to their online counterparts:

"One of the most important is that, at least for now, the quality of sound available over the Internet is not comparable to what the listener expects from the compact disc. The time required to download a sound file, with the most recent compression algorithms, is substantially more than the file's playing time. Perhaps most important, the cost differential between the hardware required to play a compact disc or cassette and the hardware needed to receive and play sound files from the Internet is enormous." (Moore 2000, 639)

Though Moore's final reason concerning the hardware cost differences cannot be argued (aside from the cost of the computer and software required to download/create MP3s, the least expensive personal digital music player available runs well over $100), it must be weighed against the ongoing issue of providing access to our users. Currently libraries have the technology to provide patrons with on-demand access to our sound recording collections from the comfort of their homes, but in most instances we are prevented from providing this service because of copyright and license issues. Once these problems have been addressed, hopefully within 2-5 years for libraries, will we continue to force users to physically enter our buildings in order to use our resources? I believe we do so at our peril.

Finally, in his essay on cataloging, Ralph Papakhian offered this insight:

"If the digital revolution does occur in music materials soon, such that the bulk of library-held notated and recorded music is converted into and stored in digital media, then we should expect that our current cataloging practices would be altered significantly." (Papakhian 2000, 588)

The first principle concerns the mode of access. MP3s are accessed much like any standard computer file, using one of two methods:

• Direct access, in which the files are stored on a physical carrier that can be easily handled by a user (i.e. a CD-ROM); and
• Remote access, where the aforementioned physical carrier is not present, and users must typically utilize a computer terminal connected to either a remote network or a local storage device in order to access the file.

Another place where the dual "sound recording/computer file" nature of MP3s is felt in terms of conventional cataloging practices occurs in the 007 (Physical Description) fixed field of the MARC record. One should make sure that the appropriate set of 007 standards is utilized. Remember that 007 fields are repeatable, as in the case of Example A (Bach's Cello Suite in G and C). This title is a commercially produced compact disc that contains both standard audio tracks as well as MP3 files. It also includes special software so users can easily access the MP3 content (provided one has a computer with a CD-ROM drive). Thus there are two 007 fields, one created to reflect a sound recording while the other one represents a computer file. Of particular note in terms of the computer file-related 007 field is the subfield $k (Level of compression). It was alluded to earlier that MP3s discard or "lose" some of the digital data as the file is compressed. This means MP3s utilize a "lossy" compression technique and should be coded "d." [Example A, line 3] Both the 500 note that illustrates the special features of this CD, and the 538 note listing the system requirements needed for the utilization of the MP3 files, are reflected in Chapter 9. [Example A, lines 10 and 14]

Chapter 9 plays an even more important role in terms of remote access MP3s. Because there is no physical carrier, the 300 field is eliminated altogether. In its place one uses a 256 (Computer File Characteristics) field, accompanied by two 538 fields (one for mode of remote access, and another for the system requirements), and finally an 856 (Electronic Location and Access) field. Of particular interest are the special "%5F" characters that appear in both the 538 and 856 fields of the remote access examples. [Example B, lines 11 and 30. Example C, lines 12 and 18] These characters are the accepted way to represent the underscore ("_") in a MARC 21 record. (Library of Congress, Network Development and MARC Standards Office 1999)

The next principle addresses the chief source of information. Both Chapter 6 and Chapter 9 cite the physical carrier (i.e. the CD or CD-ROM) as the initial chief source, followed in importance by accompanying printed material or the container in which the physical carrier is stored (i.e. a CD jewel case). This works fine in the case of a commercially printed, direct access MP3 product such as Example A. What about the chief source of information, however, for Examples B and C?

Thankfully, almost all MP3 files have something called ID3 tags. An ID3 tag contains such pertinent data as composer, title, performer, and time length, and any MP3 player can access it. The first version of ID3 tags (ID3v1) came attached to the end of files and held very minimal information. The current version of ID3 (ID3v2) appears at the beginning of the file and can hold up to 256 MB of data. This means that an MP3 file of the Hallelujah chorus from Handel's Messiah could contain in its ID3v2 tag the composer's name, the piece's collective and part title, the complete list of performers, the exact time length and file size, even the original date and location of the performance if it was from a live concert. Thus one could utilize the ID3 information in creating the bibliographic record, and then add a 500 note stating the source of the title. [Example B, line 10. Example C, line 11]

The direct and remote access examples featuring Schonberg's Klavierstuck op. 33a (1928) represent a single MP3 file that could be purchased from a Web site and then either "burned" to a CD-ROM or mounted remotely along with similar files according to the purchaser's discretion. Because the publisher did not determine the track collection, these examples could be treated as "bound-with" items. Hence, the 256 or 300 field reflects only one musical track, with the 501 field listing the other MP3s with which they are grouped. [Example C, lines 9 and 15] Subsequent catalog records can then describe each of the remaining music files in a similar fashion.

An examination of the cataloging principles most affected by the new format does not mean that other metadata standards such as Dublin Core can not be used to similar or greater efficacy. It points to the fact that we as catalogers should not wait until MP3s or other digital music formats have arrived on our collective doorsteps before deciding on a course of action. We might find ourselves in a similar situation to that of the recording industry...trying to play catch-up.

DeCarmo, L. 1999. "Safety in Numbers: a look at the Secure Digital Music Initiative." EMedia. 12/11:49-54.

Fritz, M. 1995. "Digital music organizations take first steps toward audio standards." CD-ROM Professional. 8:17.

Gorman, M., and P.W. Winkler, eds. 1988. Anglo-American Cataloguing Rules. 2nd ed., 1988 rev. Chicago:American Library Association.

Hacker, S. 2000. MP3: the definitive guide. Sebastopol, CA:O'Reilly.

Library of Congress, Network Development and MARC Standards Office. 1999. Guidelines for the Use of Field 856. http://lcweb.loc.gov/marc/856guide.html

Moore, T. 2000. "Sound recordings." Notes. 56/3:635-640.

Olson, N. ed. 1997. Cataloging Internet Resources: a manual and practical guide. 2nd ed. Dublin, OH:OCLC.

Papakhian, A. R. 2000. "Cataloging." Notes. 56/3:581-590.

Parker, D. 1999. "The RIAA and MP3: in search of a clue." EMedia Professional. 12/4:96.

Partyka, J. 1999. "Technology: the unsigned musician's best friend." EMedia. 12/11:76.

Waldrep, M. 1999. "DVD-audio: is music next?" EMedia Professional. 12/4:77.

Wright, H. S. 2000. "Technology." Notes. 56/3:591-597.

ISSN 1069-6792
Revised: 12/13/00
URL: http://wings.buffalo.edu/publications/mcjrnl/v7n2/freebornmp3.html