Apple-EMI deal dooms WMA? Huh?

Since that last post, I’ve seen a couple other articles suggesting that Apple’s decision to offer DRM-less tracks in the AAC format is going to strike the death blow for Microsoft’s Windows Media Audio (WMA) format.

Everybody loves a “Microsoft is losing” headline, but these articles are just stupid. Why?

1. If device makers had thought they could steal iPod users by supporting AAC on their devices, they would and should have done so long ago. The huge majority of AAC files on people’s computers are unprotected files from CDs ripped in iTunes, not from the iTunes Store. (This is why the Zune natively supports AAC.) Removing DRM from a miniscule percentage of the overall number of AAC tracks out there doesn’t change the equation a bit. (In fact, if I were a player manufacturer, I’d support as many formats as I could afford to license so the user would never have to think about formats at all. Isn’t the idea maximum ease of use?)
2. These writers seem to assume that the Apple-EMI deal will remain exclusive. I doubt that’s the case. EMI wants to sell tracks as broadly as possible, even to the tiny percentage of Zune users out there. Assuming they believe removing DRM will sell more tracks, why not allow Microsoft to sell DRM-free WMA files on the Zune Marketplace? And allow Napster, and Yahoo Music, and whomever else to sell DRM-free files in whatever format they choose. These folks would be happy to sell DRM-less tracks—like nearly every other digital media player, they support DRM because they have to, not because they want to. It’s costly to implement, alienates consumers, closes off interesting scenarios that might make users more prone to buy new digital media products, and so on.
3. The Business Week writer talks about Microsoft’s “expensive licensing terms” for WMA. But AAC comes with licensing terms as well! And they’re more expensive for device makers than WMA (at least according to Paul Thurott).

News flash: Apple chose to make this deal for AAC because they’ve already licensed that format, and the iTunes infrastructure is probably built to deliver files in that format, and it’s a much more efficient (filesize for quality) format than their other option, MP3. Which, by the way, comes with licensing fees as well.

The undercutting of the PlaysForSure music store partners probably did more to hurt WMA than this deal will. And Microsoft doesn’t care because those stores weren’t selling many songs anyway, and Microsoft has plenty of other ways to push the Windows Media format—native support in every PC and Windows Mobile device shipped, Xbox Live Marketplace (for video), MSN Video, Zune Marketplace and so on.

But everybody loves a “Microsoft’s doomed” headline.

WMA

Despite all of the strategic mistakes Microsoft has made in the digital media space, some of their underlying technology is pretty remarkable. In particular, the Windows Media Audio (WMA) codecs occasionally surprise and impress me.

In this context, a codec is a computer algorithm used to reduce the file size of digital media files for easier transmission over a digital network. In very basic terms, this is done by removing extraneous information from the signal–for example, bits that represent very high or low tones that most listeners cannot hear. The trick is removing the right stuff–sometimes, seemingly inaudible portions of the signal, when summed up, contribute to a particular instrument or voice’s character–remove too much, and the cymbals sound tinny, or the guitar sounds flat, or the singer’s voice sounds thin.

In the best head-to-head test I’ve seen, WMA comes out very well versus its competition.

But more sriking to me was a recent personal example. Diminished Men, for whom I’ve been playing bass, recently finished recording an album. It was done onto 2-inch analog tape, then moved into ProTools (the industry-standard digital mixing tool) for mixing. The tracks have been mixed, but not mastered. Mixing is where you listen to each individual track (bass, guitar, each drum), decide which tracks to use where (for example, we did lots of guitar takes with different tones), equalize the tracks you’re going to use, blend them together at a certain relative volume, re-equalize, re-blend, and so on. Once that’s done, you master each song, which means equalizing the whole song, compressing it, increasing the volume, and so on. This is done to bring out parts that are hidden, add power, and so on. (Mastering is also where you decide song sequence, how much space between tracks, and so on.) There’s a lot of complicated voodoo (and opinion) involved in both parts–for example, an unmastered mix might sound like crap to an untrained ear, but will sound better once the engineer has equalized certain parts up and down or compressed the whole thing.

Anyway…when I converted those mixes from full CD to 320kbps WMA files for my Zune, it actually ended up sounding like a good master. My guess is it eliminated a lot of low-level noise in the midrange (this is an instrumental surf band with lots of guitars and drums with big reverb), bringing out the low and high a bit and making the overall thing sound cleaner and crisper, while still preserving the slightly chaotic tone.

DRM deathwatch, continued

Steve Jobs has always struck me as a Zen-styled capitalist charlatan, kind of like Jerry Brown with money. Nonetheless, he has earned my enduring respect with this open letter to the music industry, in which he suggests that DRM-protected music files should be abolished. The nut: “Why would the big four music companies agree to let Apple and others distribute their music without using DRM systems to protect it? The simplest answer is because DRMs haven’t worked, and may never work, to halt music piracy.”

Shows a lot of confidence in the iPod as well–DRM lock-in is one frequently cited reason why users won’t switch. I guess Apple created FairPlay simply because they didn’t want to license another DRM system. For example, using Windows Media DRM would have required them to sell songs in the Windows Media Format, which means the iPod would have to be able to play that format, which means Apple would have had to pay Microsoft royalties on every iPod sold (up to a cap of $1m per year, which is peanuts now but probably seemed like a lot at the time).

Meanwhile, a blogger at 10 layers has looked at current earnings growth and extrapolated into a future where Apple’s revenue surpasses Microsoft’s. I say it won’t happen–it would require iPod growth to continue at its current clip (impossible) or for Apple to have one or more huge hit products (very unlikely) or for Microsoft to stop growing completely (very unlikely). Still, it’s striking that Apple’s revenue is 50% of Microsoft’s today. Unimaginable 10 years ago.