Reference

When you come to a fork in the road, take it.—Yogi Berra

Over one busy week in 1986, Karlheinz Brandenburg laid the foundation of a technology that a few years later would upend the record business. Brandenburg, a PhD student in electrical engineering at the Friedrich-Alexander University Erlangen-Nuremberg, was figuring out how to code digital music efficiently enough that it could be delivered over digital telephone lines. A patent examiner had concluded that what the application proposed was impossible, so over a week of late nights, Brandenburg produced the proof of concept and more. It was another decade before the technology—MPEG-2 level III, more commonly known as MP3—would find its true home, the Internet.

MP3 was first envisioned as a profitable business. The Fraunhofer Institute for Digital Media—one of 60-plus applied science institutes supported by the Fraunhofer Society for the Advancement of Applied Research—and resellers would sell MP3 encoders for real money. Decoders would be cheap or even free. Then, in 1997, an Australian graduate student used a stolen credit card to buy MP3-encoding software, and posted it on an FTP site with a .readme file saying, "This is freeware thanks to Fraunhofer." In the ensuing years, the illegal sharing of MP3 files via Napster and other websites decimated the record business (footnote 1).

In time, it became clear that the culprit wasn't Napster per se—or MP3, for that matter—but the Internet itself, combined with a new kind of information-based commodity. Previously, cultural artifacts had always been objects: paintings, sculptures, books, recordings. But in this new world, a song or a symphony was ephemeral; you could give it away again and again but still own it. Music was especially vulnerable because the digital code was readily available in unprotected form, buried in the pits of Compact Discs.

What's Old Is New Again
As I noted in the February 2018 issue, some critics have drawn unfavorable parallels between MP3 and Master Quality Authenticated (MQA), the music-distribution codec invented by Peter Craven and J. Robert Stuart. Those critics have drawn attention to the "lossy" nature of both MP3's compression and what MQA Ltd. calls its codec's "audio origami"—its folding of ultrasonic information into the frequency range below it, in the bits beneath the noise floor. It's a superficial comparison—and ironic, in that MQA is intended, in large part, as a solution to problems introduced by MP3 and the Internet. "We now see a plethora of file types, we see a huge percentage of the music-buying public just plain pulling out," Stuart wrote to me in an e-mail. "It is too complicated. It isn't satisfying. There are other things to do. Audiophiles still survive on the dregs and niche releases."

To fully grasp the extent of the problem, audiophiles need to look at the big picture. First, a healthy music industry is important. As Spencer Chrislu, MQA's director of content services, told me in an interview in 2016, "It's important . . . to protect the interests of studios. If a studio does their archive at 24-bit/192kHz and then uses that same file as something to sell on a hi-rez site, that is basically giving away the crown jewels upon which their entire business is based." The metaphor seems mostly apt, although crown jewels have the profound advantage of being physical objects; they can't be given away and still owned. Plus, jewels are harder to alter than a music file.

Another big-picture item: Audiophiles need to recognize that we're a small minority among music consumers. When have our interests and opinions influenced any high-level decision in the music industry? The cases I can think of were all eventual failures: HDCD, SACD, DVD-Audio. The best we can hope for is a system designed to serve the interests of others—the industry, musicians, and casual (and mobile) music listeners—but that is also good enough that we can live with it.

With MQA, record companies can supply consumers with versions of recorded music that sound at least as good as the best-quality master recordings in their archives—MQA claims that they sound better—without sharing actual digital masters. An era of music distribution dominated by MQA begins to seem like a sort of post-apocalypse version of the peak era of vinyl, when music was issued almost exclusively on black discs (there probably were some open-reel tapes around), which sounded fine, or not so good, or great, depending on many factors. Those legendary master tapes were kept hidden away in musty vaults (footnote 2). MQA proffers a simpler world in which a single, back-compatible distribution format serves all needs, and in which consumers no longer have access to those high-resolution PCM masters.

That, anyway, is the vision of MQA Ltd.

To most audiophiles who experienced the pre-Internet music world—that is, most of us—that vision is pretty appealing (though I do like my 24/192 downloads). MP3, loudness wars, earbuds, Beats, convenience over quality, dead record stores, playback rituals destroyed—has anything good happened in music technology since, say, the Replacements broke up? Buying online, downloading, and managing a digital music library are not nearly as much fun as a weekly expedition to the neighborhood record store. Few 21st-century experiences can match that of carefully pulling a new prize from its paper sleeve, putting it on the turntable, brushing off the dust, and lowering the needle.

Regrettably, it's hard to envision a future that includes bricks-and-mortar record stores. But perhaps some sanity—and some profitability—can be restored to music production and distribution. Sure, we've always known that the major labels cared more about profits than about music, musicians, and us. Without them, though, we'd have missed out on those wonderful experiences. Surely, a revivified music industry would be good for people who care about music, even if we audiophiles remain an afterthought.

Lurking just below the surface of the argument I'm making is a vulnerability—or an entire class of vulnerabilities. I'm looking backward. So, in a way, is MQA.

I'm somewhat, if not yet entirely, convinced that MQA is based on a real, advanced, sophisticated technology. It's optimized for streaming, the dominant music-delivery system of our time for better or, more likely, worse. MQA's ideas about time-domain performance and its origins in post-Shannon sampling theory reflect cutting-edge thinking. I find its focus on neuroscience less convincing, but that, too, makes a compelling story. In many important ways, MQA looks forward.

And yet—MQA is a locked-up, proprietary technology in a world that, influenced by information technology, has come to expect open sources and standards. And that locked-up technology is aimed, apparently, at restoring an old-fashioned, label-dominated musical economy—which is surely why the three major record companies are all now stockholders. Even MQA's notion of "analog to analog"—microphone feed to DAC output—can seem old-fashioned in assuming that music starts its life as analog in an increasingly digital world. Not all music comes into existence as vibrating air, and even when it does, it often matures largely in the digital realm.

Audio Files for Audiophiles
Napster and MP3 aren't the only changes wrought by computer technology. The growing Internet has swallowed ever-larger slices of culture and commerce, employing ever-expanding hordes of IT workers. Traditional audiophiles may not have noticed, but somewhere along the line, people who listen to music via computers have come to outnumber those of us who listen through serious home audio systems, the ranks of which have shrunk over the same period.

---

Footnote 1: See The Appetite for Self Destruction: the Spectacular Crash of the Record Industry in the Digital Age, by Steve Knopper, Free Press, 2009.—John Atkinson

Footnote 2 . . . where, unfortunately, they were all too often lost or damaged by floods, neglect, or time.—Jim Austin