Pop culture obsessives writing for the pop culture obsessed.
Pop culture obsessives writing for the pop culture obsessed.

Ask The A.V. Club: April 13, 2007

Youth Culture Crashed My Computer

I heard a rumor that a computer program is being designed that will "listen" to a song and determine how popular/profitable it will be, based on a variety of factors. Is this even remotely good for humankind? Or am I overreacting, since "focus groups" are kind of the same thing and are used already?



Donna Bowman responds:

Remotely good for humankind? Of course not, but neither are trans fats or Thomas Kinkade. Since when does what makes good business sense have anything to do with what's good for humankind?


The computer program you mention—well, the one I know about; there probably are others—is called Platinum Blue. Malcolm Gladwell, author of Blink and The Tipping Point, wrote about it in the October 10, 2006 issue of the New Yorker. (The piece, "The Formula," is now available in Gladwell's online archive.)

The principle is that songs that become hits have distinct clusters of structural components (melody, tempo, rhythm, harmony, key, and so forth. There aren't just a few of these clusters—there are 60, according to the program's analysis. The people who run the service expect to be hired by record labels, who will ask them to run new albums through the software, looking for a single that has hit potential, as demonstrated by its location in or proximity to one of these hit clusters. Or, in a business model slightly more threatening to artistic purity, the label will run songs through the machine, then ask Platinum Blue for notes.

Say a song needs to be speeded up, or needs more bass, or a different sonic frequency, to move it closer to a hit cluster. Well, the label can take that information back to the producer, and ask for changes. This is quite a bit different from focus groups, which have the shortcoming of encouraging groupthink, and are in any case far from statistically significant. What you want a computer for is to take the human element out of the decision-making.


We don't know why we like something. Even the most talented A&R professional doesn't know whether we'll like something. That's why the vast majority of records put out by major labels fail to score a hit. But if hits have something in common, and a computer can find it, then a record can improve its chances in the marketplace. Of course, maybe you as a consumer, Ian, don't care if a record becomes popular—you just want to listen to music you like. Maybe, like our own Steve Hyden, you have reasons to hope that the music you like doesn't become popular, because that will cause you to like it less. But there's something to be said for the pursuit of popularity in the music business. A song you might like has a lot less chance of entering your orbit if it doesn't achieve some popularity. And all of us respond, at some level, to mass-media phenomena—Gnarls Barkley's "Crazy," for instance, which the Platinum Blue system correctly predicted would be a massive, culture-saturating hit, turned out to be the song we were all humming in the summer of 2006. Hits are about labels and distributors making a lot of money, of course. But they're also about the creation of a shared culture that unites music listeners across race, class, and region—and many of us would argue that such a pursuit isn't all bad, even if we didn't want all culture to be reduced to the search for massive popularity. And not all culture is likely to be. Some musicians and labels actively seek hits, some are indifferent to the prospect of popularity until it might be thrust upon them, and some are repelled by the idea. Some consumers want to have their tastes validated by the masses, some aren't inclined to seek beyond what's in their immediate environment or cultural niche, and some seek out the unpopular, repulsed by lowest-common-denominator tastes. Isn't there room in music for all of us?

Another reason to be concerned about a machine influencing what songs get released (and how they get tweaked) is an opposition to manufactured music, a kind of assembly-line process that works by rote, according to formula. Doesn't sound like art at all, does it? But how much of what gets popular is really the familiar recycled? Musical movements come and go with increasing speed. The counter-argument that the pop charts are committed to novelty at the expense of all other artistic values has plenty of support. Platinum Blue calculates a "periodicity grade" that is intended to show how "now" the sound is, how likely to connect at this given ephemeral moment—because, explains Gladwell, in any given month, only 15 to 20 percent of the hit clusters Platinum Blue identifies are "active," or uppermost in the public consciousness. Surely few would argue that "Crazy" recycled a mainstream, familiar sound, and novelty hits and quick fads have longed reigned on the charts, right alongside more enduring genres and nostalgia. Platinum Blue (and Gladwell's article) raises a host of fascinating questions about the relationship between art and commerce, and between artists and their tools. In the end, it's how those tools are used that will determine how good they are for humankind.


Wherefore Art (Thou) Cinema?

There are many movies that I want to see but am unable to because of a variety of reasons, including unavailability (like Love Streams) or availability only on extremely poor-quality VHS (most films by Buñuel, Mizoguchi, and Ozu, for example). How is it possible to watch these movies in decent quality without waiting for Criterion or Kino to update their catalogues? Are there other companies that release classic and foreign movies on DVD? The only other one I am familiar with is Image Entertainment, but the quality of their transfers isn't as good as Criterion or Kino. I live in a small town and cannot easily get to New York, Boston, or San Francisco, where they are most likely to show these films on a larger screen.


Zachary B Zimmer

Fellow landlocked cineaste Noel Murray responds:

Well Zachary, you can't go wrong with Criterion and Kino, though I wish the latter made more of an effort to provide the kind of context-setting bonus materials—like interviews, commentaries, vintage shorts, and TV programs—that Criterion digs up. But Kino covers a lot of ground that Criterion doesn't, both in terms of vintage art films and recent highbrow foreign fare, which makes them invaluable. You should also check out Criterion's "sister company" Home Vision Entertainment, which is under the Image umbrella. Since Image bought it out, HVE hasn't had as many splashy releases—at least nothing on the order of the Mikey & Nicky or Kinji Fukasaku sets that got so much attention three years ago—but it still handles some choice American cult films and recent foreign oddities, if you can navigate the Image site and find out what they are. Ranging even further into obscurity, Facets DVD routinely unearths the kind of older foreign films that only show up at big-city rep houses, alongside the work of well-regarded cult auteurs like Bela Tarr. In the past, Facets has had a reputation for shoddy presentation and negligible promotion, but they've been improving.


The offerings from labels like Home Vision and Facets tend to be a little esoteric, so if you're looking for foreign films with more crossover appeal, you should consider the output of boutique labels like Blue Underground, NoShame, Cult Epics, and Tartan's Asia Extreme. That's where you'll find a lot of new and old grindhouse fare, from Europe and the Pacific Rim, as well as—in NoShame's case, especially—a few neglected masterpieces of world cinema. And don't forget the growing number of companies like ThinkFilm, IFC, and Magnolia, which release festival favorites theatrically, then get the DVDs on the market quickly, in nice packages. Outside of Criterion, you can count on a pretty mixed bag from all these companies, but if you haunt their websites, read descriptions of the titles, pick what you like, and then subscribe to a service like Netflix or GreenCine, you can make some real discoveries.

So that's my official answer. Now come with me into the corner as I speak to you in a whisper about some other alternatives. No, not BitTorrent sites—though they're out there, and many of my friends get a lot of good use out of them—but "grey market" bootleg sites, like 5 Minutes To Live and Super Happy Fun, which make out-of-print and rare material available on middling-to-low-quality DVD-Rs, at reasonably low prices. You can also find good deals on region-free Asian DVDs at too many sites to mention, if you want to catch up on the Kim Ki-Duk and Takashi Miike films that haven't been released stateside. And did you know that YouTube has pieces of a lot of classic foreign films up for viewing—for now, at least? Take a chance by searching for a title. You never know. (Or start with the Subterranean Cinema site, which rounds up some of the more interesting underground links.)


Bottom line? No matter where you are in the world, if you have a computer and a mailing address, it's possible to be a well-informed film buff. Welcome to the club.

A Dancer On The Roof

Way back in elementary school in the early '80s, we were forced to watch some bizarre stuff: the filmstrip versions of "Danse Macabre" and "Night On Bald Mountain," Free To Be You And Me, and many other craptacular wastes of time, while the teachers took smoke breaks in the lounge. There was an animated short film (maybe 20 or 30 minutes long) we used to watch about a giant who lived alone in an old, dark house. He would watch the children play in his yard (but not in a creepy way). I also remember a Jack Frost-type character who would dance on his roof when winter arrived. Any idea what the title of this movie is? It was strangely fascinating.


Kevin Palka

Tasha Robinson apparently wasted bits of her childhood too:

You're certainly thinking of an animated adaptation of Oscar Wilde's short story "The Selfish Giant," an overtly Christian fable in which a group of neighborhood children play in a giant's garden until he returns from a trip and kicks them all out. Once the children are banned from the garden, spring and summer start avoiding it too, leaving the giant stuck in year-round winter until he learns to be less of a crabby asshole. You can read the complete text of the story here. It's been adapted to animation multiple times throughout the years, but based on the length of the piece you describe, and the specific image of the character dancing on the roof, I'm going to guess you're thinking of the 1971 Canadian version, which you can watch in segments on YouTube. I'm pretty sure I saw the same version when I was a kid—I particularly remember Hail dancing around on the roof too—though I don't remember the song at the beginning at all. Maybe it was excised from the school-distribution version to shorten the run time a bit.


Next week: Is American Idol ruining music for future generations? Also, a question of special credit, plus the most literary edition of "Stumped!" ever. Send your questions to asktheavclub@theonion.com.

Share This Story

Get our newsletter