February 1998
Mars Needs Windscreens!
Formats in Space

by PAUL D. LEHRMAN
ILLUSTRATION: STU SUCHIT

I've been working for the past few weeks on the score for a cable documentary dealing with the issue of life on Mars. It's a pretty exciting topic and a pretty exciting time to be dealing with it, what with all the recent and upcoming probes and landers. And I'm having a lot of fun writing a score that's equal parts Enya, Gustav Holst, Pink Floyd and Mussorgsky.

One of the segments of the film describes a "time capsule," created by The Planetary Society and Time Warner, which accompanied a Russian Mars probe launched in 1996. Like most time capsules, it was designed to be "opened" some decades or centuries from now, hopefully by humans who have colonized Mars. It contains artifacts of our current civilization, including essays and science-fiction stories about Mars, several generations of artists' conceptions of the red planet and...audio clips.

Unlike most time capsules, however, these were not left on the Martian surface in the form of discrete objects, like newspapers, books or films. Certainly, no one was going to pay to bury them in the Martian soil, and, left on the surface, even in Mars' noncorrosive atmosphere, they would deteriorate quickly. Besides, the weight would add considerably to the expense of an already costly project.

No, all of these items were coded onto a CD-ROM, which was attached to the leg of the probe. Presumably, 20 or 50 years from now, the new Martians, our descendants, would find the CD-ROM called "Visions of Mars" and view its contents, to their edification and delight.

If this rings a bell, you may be recalling the Voyager interstellar probes of the late '70s, the first man-made objects to leave the solar system. Inside the probes was a phonograph record called Sounds of Earth. Except for the fact that it was made out of solid gold, was recorded at 162/3 rpm (to cram as much as possible on there), and included about 100 black-and-white and color photos encoded into analog, it was an ordinary LP. Any civilization that discovered it a thousand, or a billion, years hence would presumably be able to look at the grooves with a microscope and deduce that these were sound waves, and then be able to build a device that would allow them to listen to it. (It would still be a lot smaller a creative leap than when Jodie Foster, in the movie Contact, realized that the blueprints being beamed down from the cosmos looked so weird because they were actually three-dimensional.)

Of course, the finders of Voyager may not know exactly how fast to spin the record, but as composer Laurie Spiegel says, "We don't know how fast they're listening." No doubt they will play it at whatever speed they find comfortable to listen to, wherever their hearing (or whatever equivalent sense they have) falls, in the range from DC to X rays. Some of the recordings are mono and some are stereo, and it's fun to speculate on how extraterrestrial listeners might interpret this fact--perhaps, as the late Carl Sagan, executive producer of the project, hypothesized, they will conclude that some of the Earth species who contributed to the project have only one ear. (No, there's no Beach Boys on it, and shame on you for thinking that!)

Spiegel is responsible for the first musical piece on the record, a realization of Johannes Kepler's treatise on the movement of the planets, "Harmonia Mundi," using a custom FORTRAN program that fed Kepler's data to Bell Labs' experimental GROOVE synthesis system. Besides being gorgeous, the piece could allow a clever civilization to figure out where the craft came from by examining the data and comparing it to all of the solar systems in the galaxy. Other sounds on the record include the music of Bach, Mozart, Chuck Berry and Louis Armstrong, folk music from a dozen different countries, nature recordings such as birds and volcanoes, whale songs, and spoken greetings in 55 languages. The audio clips are all from professional sources like the famed Elektra Sound Effects Library, and the brilliant field recordings of folklorist Alan Lomax.

Warner New Media (now Time Warner Interactive) released the record on a hybrid CD-ROM in 1992, so you can listen to what they'll be hearing a million years from now on Rigel 4. It's quite a wonderful document, and it makes you feel pretty good about our planet and our species.

Okay, back to Mars. The Mars CD was not meant to be found by other civilizations; it was meant to be found by humans. And our technology has advanced quite a bit in the ensuing two decades, so one assumes we could find more clever ways to present information and cram more of it into one place. "Visions of Mars" is a presentation, done with Director or Premiere or something similar, with onboard player software for use in either Mac or Windows computers.

It's not the most effective use of the medium I've ever seen. There's no video (although apparently some videotapes were made for the project), the pictures are pretty grainy for the most part and look lousy if your monitor has more than 256 colors, and it's slow. Some of the text is in a badly kerned, hard to read, Avant Garde typeface, while other texts run off the screen. The data on the disk takes up only about 160 megabytes, one-fourth of what could be on there.

But beyond the merits, or lack of same, of this particular disk is a more troubling question: Isn't it a bit of a stretch to think that CD-ROM will be a viable medium 20, 30 or 100 years from now? If you think about it, the only medium that we have today that we can guarantee will be readable in future generations is the printed page. The more technology you need to interpret a document, the less likely it is that a person at some random point in time and space will be able to read it. Looking at a phonograph record under a high-power lens could reveal to an intelligent being the nature of the information it contains, but could the same be said for the almost random pit patterns of a CD?

And how is someone in 2098 going to deal with the fact that they need to have something called a "Mac" or a "Windows PC," and it needs to have on it something called an "operating system," which must be of early '90s vintage? (How much you want to bet this won't run under Windows 2001?) I have Apple II software that's ten years old, and I'm hard-pressed to find a machine that can even read the disks, to say nothing of run the programs on them. To anyone in our industry who cares about either the past or the future, this issue is of paramount importance: how to preserve what we've done in our 100-year history and make it accessible to audiences 100 years from now.

Unfortunately, there's another issue, even more dismaying, with this CD. The producer of the documentary I've been doing is including in the film some audio extracts (and still pictures) from the CD of famous people saying hello to the future Martians. When I heard his first temp audio mix, I was surprised at how awful the clips sounded. I asked how he got them, and he replied that he played the CD on his Power Mac, plugging the audio output of the computer right into his editing system. "Pshaw!" I countered. "I have many sophisticated tools for extracting and converting audio from CD-ROMs, and if you send me the CD, I will create a 44.1kHz, 16-bit file of the clips, which will sound perfect, that you can load right into your Avid."

He sent me the CD, and I hacked at it for the better part of a day before I gave up. The files were in some weird format that I couldn't get to play in any of the dozen or more audio programs I threw at them. The closest I came was with Tom Erbe's amazing SoundHack, which can make sounds out of any file format, from MPEG to a Photoshop document. It got the files open and started to play them, but after a few seconds they dissolved into a harsh buzz. So I was reduced to using, like my producer, the audio output from my Mac. I figured I could still do him one better by routing the analog signal through the A/D converters on my high-end DAT machine, but when I tried that, my dismay gave way to horror: No matter how I got to them or what I did to them, the audio files were just plain garbage.

To get Arthur Clarke on tape, they must have sent him the cheapest portable cassette recorder they could find, and they forgot to tell him that he shouldn't record himself sitting in his garden during a typhoon. Carl Sagan's voice is distorted and full of cracks and pops, and children's book author Judith Merrill sounds as if she's at the bottom of a well. The recordings made me recall the time I captured my three-year-old sister doing her Shirley Temple imitation on my father's dictation machine. I was six. My tape sounded better.

I sent the CD back to the producer and told him that I failed, and he had already done the best that could be done. During the final mix process, some of the scratches and crunches on Carl could be painstakingly excised, and some radical EQ might make Judith a little more intelligible. The wind noise on Arthur was impossible to deal with, however, and so the producer would simply salvage what he could between gusts. We shook our heads and laughed over how such an ambitious project, with such a lofty goal, could be done so badly.

It's okay for the best minds in our industry to be squabbling about whether analog mastering sounds better than digital. It's okay for us to argue over the merits of oxygen-free cable, tube preamps, oversampling or active crossovers, or to debate how many channels of audio should be dancing on the head of a DVD, at how many bits per word and how many words per second. But Visions of Mars, like too many products of its ilk, forcibly reminds us that we need to band together to combat a very real enemy: the people who make the multimedia today and who are building the delivery systems of tomorrow--and I'm not talking about space probes--who know nothing, and care less, about how things sound.

That an independently produced 20-year-old record, re-released on CD, should be far more successful--technically, aesthetically, educationally, and in terms of the amount and value of the information it presents--than a modern, state-of-the-art interactive presentation produced by the world's largest media company, should give us all pause.

Fortunately, "Visions of Mars" never made it to Mars. The Russian Mars probe, like all other Russian Mars probes, never got there. It's probably drifting around somewhere in the Asteroid Belt by now. Future Martians, if there are any, will probably only encounter it in the cutout bins at Egghead. Then they'll have to go over to Fry's Antiques to find something to play it on. And when they've gone to all that trouble, they'll marvel at how we could have been so stupid as to want to leave this as a legacy for the future.


Update department: A couple of months ago I wrote that I had abandoned the MIDI sequencing program I had been teaching with for ten years, in favor of a newer, slicker program. It turns out it was a mistake. As much as I liked the new program (and no, don't ask me what it was...I would like to stay on speaking terms with the manufacturer), it suffered from the very malady that I described in that column, a malady that all technology-oriented educational programs are in danger of falling into: the inability to look far enough into the past to learn why we do the things we do.

In this case, the program ignored a decade or more of user-interface development and locked users into its own, rather peculiar way of doing things. Worse, a lot of features that appealed to me upon first encounter turned out to be unfinished, and unusable, while the simple operations one needs in a teaching program were made unnecessarily complex. I struggled with the program for six weeks, watching my students suffer far more than they should (though I do believe in suffering for your art, there's a limit), and before the semester got too far under way and I found myself marooned in software hell, I pulled the plug and went back to my old sequencer.

Most of the students were grateful, and discovered that despite missing a few of the new program's cooler features, they could work a heck of a lot faster once things began doing what they were supposed to. The old program is buggy (one real whopper is that if you are running a SMPTE source into the program while you launch it, it crashes immediately), but I know where the bodies are buried, and so I am able to give the students plenty of tips about how to work around the bugs. The lesson to be learned is that the devil you know is preferable to the one you don't know--which, of course, is a primary reason why incumbency is the most valuable asset a politican has. But it also illustrates that those of us in the development biz must be forced to come up with much more concrete ways of convincing people to change their work habits and use our wonderful new toys than the usual smoke and mirrors...but that's another column.


Paul D. Lehrman has both feet planted on Earth, although other parts of him are in cyberspace.

These materials copyright ©1998, 2001 by Paul D. Lehrman and Intertec Publishing