by Rich Wilson © 1999 All Rights Reserved
This article began its existence as the May 1999 Philadelphia ITVA meeting report. A custom writing algorithm in the author's brain compressed that article to fit in the June mini-newsletter. Then it was decompressed, and proprietary enhancement techniques yielded the new, improved version you see here. Readers may note that a tongue-in-cheek lead paragraph is an unavoidable artifact of the compression/expansion process.
The focus in May was digital media compression, with an emphasis on MPEG-1, Layer 3 audio, commonly referred to as "MP3." Brad Litwin, the organizer and main presenter, pointed out that MP3 is not revolutionary per se. In fact, the specification is already more than 10 years old. So why is MP3 such a hot topic all of a sudden? A quick review may help.
Many readers know that video
and audio, when converted to digital form, result in huge amounts of data.
Telephone wiring, still the most prevalent point-to-point communication system
in the world, can't handle that much data quickly enough. So people have long
researched various codecs (code/decode methods) to shrink the original images
or sounds for faster transmission, then expand it again at the receiving end,
all while keeping the best quality possible.
MPEG (Moving Pictures Expert Group) is the nickname for a subgroup of the International Standards Organization (ISO). It set out to create standards for video and audio codecs, and MP3 is just one of many variations arising from that effort. Simultaneously, many private companies went their own ways, developing proprietary codecs, some of which surpass MP3 in quality and efficiency. Some became well-known on the Internet, particularly RealAudio from Real Networks, which was designed for real-time "streaming" delivery.
Since it was designed to fit through a very narrow "pipeline," most early versions of RealAudio were of poor sound quality, and the cost of RealAudio web server software slowed its growth. Meanwhile, competing commercial products ran into various other roadblocks. All in all, private industry was slow in bringing a really good audio compression method to the marketplace.
End users finally took matters into their own hands. Many decided that they were willing to give up real-time, streaming audio in favor of better sound quality. For those willing to wait a few minutes for a file download, MP3 was the best answer available. It's efficient, it's cheap, and it sounds reasonably good.
How good? Side-by-side, most people can readily distinguish between an MP3 file and the same content played from an uncompressed Compact Disc original. This was demonstrated by an A/B comparison at the May meeting. But if one only hears the MP3 version, without the comparison, one might not notice the missing quality, particularly in a car, dorm room, or the like. In short, it's not perfect, but for most purposes one would say it's "good enough." More to the point, an MP3 file sounds "good enough" even though it is only a fraction of the size of the original.
And the MP3 codec has one big factor in its favor: while some patents have arisen from the implementation of MP3, the standard itself is in the public domain. Most competing compression standards are proprietary to commercial interests. So any clever person can write an MP3 encoding program without paying a licensing fee to anybody. Various people have written such programs, and often distribute them at low cost or for free.
MP3 has found two primary applications. In one, music fans encode their favorite songs from their own collections, using the CD players in their PCS. They then either play the music back over their multimedia computers, or download the MP3 files to portable solid-state players like the Diamond Rio. These players have few moving parts, unlike disc or tape players. This makes them skip-proof, and theoretically more durable. Consumer demand has already brought prices for portable MP3 players from $200 down to $150, and one assumes they will get cheaper yet.
The May ITVA meeting
focused on the other main niche being filled by MP3: enabling musicians and
composers to publicize their work quickly and cheaply. Besides his interactive
CD-ROM design work, Brad Litwin is an accomplished blues guitarist and singer.
He sees MP3 as a means of leveling the playing field for talented but
lesser-known musicians, many of whom would go unnoticed by the huge companies
now dominating the entertainment fields. He demonstrated a website
appropriately called MP3.com, which provides free web pages for musicians.
Visitors to a typical artist's page there can listen to a song immediately using a RealAudio streaming audio "plug-in" with a web browser; they can download individual MP3-compressed songs for later playback; they can order a CD compilation of songs by the artist, for which the web site makes all the money, or the visitor can order a "DAM" CD, on which the web site shares some of the money with the artist. The artist's page may also contain links to other places; in Brad's case, a visitor could jump to Brad's own web site, and order a true high-fidelity CD from him directly.
Brad showed the pages of statistics that the MP3.com web site provides, and as one might guess, many visitors download MP3 songs and enjoy Brad's music, while Brad doesn't make a penny from it. For now, he seems content that he is reaching a larger audience. And apparently, many others feel the same way, because each day more musicians upload their works and create new pages on the MP3.com site. Major artists like Tori Amos and Alanis Morisette are using this free distribution method as a way of promoting interest in their other work, although it's a safe bet that their record companies would never condone giving away entire albums for free.
Many people at the meeting wondered about an artist's lack of control over copyright, when so many computer users can easily download software that "rips" audio CDS (turns CD music tracks into MP3 computer files), then share those files without permission from the artists or their record companies. Brad even demonstrated how easy it was to do this.
Jon Cooper, Director of
CD-ROM Business Development at DiscMakers, was next to speak. He discussed the
"business model" of free music delivery using MP3 technology,
pointing out that it undercut the revenue stream of the traditional music
industry. He observed that major music retailers (like Tower Records) stick
with RealAudio on their web sites. It is faster, although slightly lower in
quality, and gives customers a preview of an artist's music, thus facilitating
traditional mail-order sales of commercially-produced, high-quality audio CDS.
However, the explosion of pirate web sites with MP3 copies of commercial
recordings is forcing retailers and record companies to shift their plans and
rethink their use of technology.
Jon cited "Liquid Audio," an alternative method of compressing and "streaming" audio in real time, which allows the owner more control over music licensing and revenue. Another advantage is that Liquid Audio files are automatically scaled to fit the bandwidth available at the time of delivery, unlike any other compression format. Despite its advantages, Liquid Audio has yet to catch on, due to its somewhat more complex installation requirements for users and potential customers, among other things. However, new Liquid Audio versions continue to evolve and add new features, including the ability to play RealAudio files.
There are many
other audio codecs and algorithms out there; AAC, (Advanced Audio Coding) is
part of the MPEG-2 specification. AC3, a product of Dolby Labs, is generally
associated with its application in surround sound for motion pictures, but
Dolby hopes to see it used in television and video production as well. MPEG-4,
which is used in Digital Television broadcasting, is a more advanced system
that not only allows interactivity, but also allows greater protection for
intellectual property rights (i.e. copyright), compared to other MPEG formats.
The current momentum and popularity of MP3 might carry it past other, superior compression schemes, as has happened in other technological fields. People at the meeting immediately compared it to the VHS-Betamax history. Longtime ITVA members who have lived reluctantly with the quality constraints of the NTSC video specification -- and waited many years for a viable HDTV system -- will recognize a similar pattern there, too. So, although MP3 is not the best-quality audio codec available, it may become the de facto standard.
Jon concluded by suggesting that the current proliferation of MP3 might encourage something he saw as very positive: the popularization of "micropayments." These are a form of Internet "scrip" that users could agree to pay, using a secure browser connection to a web site, in small increments from a pre-established credit or debit account. ("Cybercash" is an example of a company set up to handle such payments.)
If such a system were in place, users might be very willing to pay 5 or 10 cents to download a copy of a song, for example. It doesn't seem like much money by itself, which encourages the user to spend it readily; but it really adds up for the artist, when one considers the high volume of visitors to some of the MP3 web sites. In such an arrangement, artists, distributors, manufacturers and retailers would take on radically different roles, and huge entertainment companies might lose their current total dominion over music industry revenues.
Naturally, some non-musical ITVA members wondered how MP3 would
affect their work. The best example was provided by Gary Moskowitz, owner of
Baker Sound, who helped sponsor the meeting. Baker Sound already has the
ability to hook up with studios all over the country using ISDN, a telephone
technology that allows the exchange of full-spectrum audio in real time. But
there are many studios that lack such connections.
When recently faced with the need to send audio files to a distant client on a same-day basis, with no ISDN available, Gary first used MP3 compression to shrink the edited digital sound file (a typical MP3 compression ratio is about 10-to-1). He then sent the MP3 file as an attachment on an e-mail message to the client. It took the client just a few minutes to download the sound file, "decode" it into a format that could be dropped into an editing system, and go to work.
Gary's examples dealt primarily with radio spots and similar work, but he felt it wouldn't be unreasonable for someone to use MP3-compressed files in other emergency situations, especially for voice-only tracks. It would allow long-distance collaboration on video programs when there simply isn't time for an overnight express shipment of a traditional audiotape.
Your correspondent found the May meeting inspirational. Armed with so much new information, I went home and retooled my website. I have had voice sample files posted online for some time, but they were in native (uncompressed) file formats like WAV and AIFF. I went back to the same source material, digitized it at higher sample rates, then compressed the resulting files with an MP3 encoder.
The MP3 versions are only one-third to one-half the size of the older WAV files, but sound virtually identical, especially on a typical multi-media computer. I copied some of the files to analog cassette and played them over my home stereo, and found them quite decent even at higher volume over good speakers. Considering that they take less than half the time to download than the older files, that's a real help to users and potential clients.
As many ITVA
members know, the MPEG program also developed video compression algorithms,
besides the audio compression that dominated discussion at May's meeting. And
the Internet continues to influence the direction of future research and
development in these areas. So the technical and business implications of the
MP3 example may be worth noting, as available bandwidth increases, and
streaming video becomes more viable. Will impatient consumers settle for
Internet video that is "good enough," instead of demanding the
"best" quality? Time will tell.
Rich Wilson is a freelance writer, actor/narrator, producer and consultant, and four-time winner of Philadelphia ITVA's Craft Award for Voice-over. He has set up a web page where readers can explore MP3 and related topics, at http://www.online-communicator.com/demos.html
