Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
I, Cringely - The Survival of the Nerdiest with Robert X. Cringely
Search I,Cringely:

The Pulpit
The Pulpit

<< [ The Sweet Spot ]   |  Lock and Load  |   [ Tough Love ] >>

Weekly Column

Lock and Load: Apple's Rumored Storage Locker Strategy Would Allow Us to Look But Not Touch

Status: [CLOSED]
By Robert X. Cringely

Two interesting events took place while I was distracted with Google over the past couple weeks -- events that give us just a bit more information about where the future of digital media distribution is likely to end up, and the risks companies are willing to take to be part of that future.

The first event was's announcement of Oboe, its music locker service. MP3tunes is one of the companies owned by Michael Robertson, who founded back in the Internet boom years and sold it for a ton of money. Robertson, a serial entrepreneur, has Linspire (formerly Lindows) and SIPphone and probably other companies, too, in addition to MP3tunes. Daring and willing to spend money if he has to, Robertson is a force to be reckoned with.

The Oboe music locker is a simple concept: For $39.95 per year, you can upload your entire music collection to an MP3tunes server and listen to it anywhere in the world you can get an Internet connection. There are no storage limitations or extra bandwidth fees. The system is legal because the MP3tunes server can only be accessed by you, so it is no different than ripping your CDs on your home PC. Of course, you have to bother to do the uploading, but special software makes that fairly painless if lengthy. Music is encoded at a maximum of 192 kbps, so this isn't really DVD-quality, but I don't generally get along well with the kind of people who can hear the difference, so it doesn't matter to me.

I just wonder how Oboe can be a profitable business? I can see it from a bandwidth standpoint. My bandwidth cost for NerdTV, for example, works out to about $0.16 per downloaded gigabyte. Streaming at 192 kbps that means your $39.95 could pay for 2,894 hours of music, which is far more than your average user will play in a year. But bandwidth is probably only a percentage of the total cost for creating and marketing such a system. If raw bandwidth is only 25 percent of the cost and the average user listens less than 700 hours per year Oboe can make a profit, but I have to wonder about the storage and infrastructure costs? If they really keep each user's music unique, they'll need petabytes of storage. Can this business really make enough to be worthwhile or is it just a Trojan horse for something else?

So I asked Michael Robertson.

"Don't look at bandwidth and storage economics today," said Robertson. "Instead ask yourself what the economics are tomorrow. This is not a Trojan horse. There will be millions of music lockers in five years and it will a profitable delivery business. Think cable/satellite versus ABC/NBC. We will be the cable/satellite for the delivery for music."

And isn't Oboe likely to be sued by the record companies?

"If they sue me, they should sue Gmail first," says Robertson. "Gmail gives you two gigabytes of free storage with no access controls -- nothing stopping users from signing up for anonymous account and then loading that up with 2 gigs of music then widely distributing the password. Of course, the same is true for every ISP and general-purpose locker system like .mac, streamload,, etc. If the music industry sues, I will politely point them to the DMCA."

Still scratching my head, I'm not sure if I buy the economic argument. Yes, storage and bandwidth costs are going down all the time, but Oboe still has to function in the real world and I think to be profitable ,Robertson will have to hope that most subscribers don't upload very much music. A few gigabytes are fine, but Andy Hertzfeld's terabyte of digital music (including every live Dylan concert ever recorded) would be unprofitable for Oboe today OR five years from now.

But Robertson is better at making money than I am, so this idea of an accessible media storage locker will probably stick.

Apple Computer evidently thinks so. The second important event I alluded to above has yet to actually happen, but the Mac rumor site seems quite sure that Apple will announce a video locker strategy of sorts at the January MacWorld show.

Though I have written quite a bit about what I believe to be Apple's emerging media strategy, I have no specific knowledge about this upcoming announcement. That doesn't mean, however, that there isn't a lot of material to analyze from just the Thinksecret account, which you'll find among this week's links.

Apple's take on enhanced Digital Rights Management, according to Thinksecret, is never actually allowing copyrighted bits to be stored on the user's machine. Instead, when you buy or rent a movie from iTunes, the video will sit on Apple's server and be accessible to the purchaser or renter in accordance with the rules of that particular transaction. If you don't have the bits on your machine, the theory goes, you can't steal them.

Yes and no.

It is a clever plan and one that actually makes a lot of economic sense because Apple isn't saddled with Oboe's task of keeping five million different copies of Louie-Louie on its server. The Apple system can keep only a few copies and simply assign access rights. For storage, Apple's reported plan is a winner, then. It works well for bandwidth, too, in part because the data only ever travels in one direction, unlike Oboe, which has to pay to receive it from the subscriber then pay again to serve it back.

I'm not sure it works as well from a security standpoint. There are already software products that can hijack video and audio streams in real time as they play and though Apple can work to defeat those, it will probably be a losing cause. Still, with FairPlay on top of this locker strategy, Apple has something better than what's being offered by its competition. Movie studios are eager for new revenue sources and Apple has the mojo, so we'll probably see a bunch of movie studio announcements to this effect in January.

But we're not there yet. What really bothers me about this Thinksecret item is how they finesse the language of what's actually happening from a technical standpoint. Yes (if the item is correct at all) the movies will stay on a server somewhere. Yes, they'll be played on your PC or Mac. And yes, they'll never be stored there because of some clever Apple caching technology. But what does all this come down to, really? Streaming.

For the Thinksecret item to be correct, Apple will be in the business of streaming movies rather than downloading them.

Downloading is easy, streaming is hard. Downloading can take as long as required with the only real criteria being that at the end of the process a perfect copy is on your computer. Downloading can work over any speed of connection and can be made to work around any other network traffic. Downloading is slow but polite.

Streaming, on the other hand, is demanding. Streaming likes to have as much bandwidth as possible RIGHT NOW. To make streaming of high-bandwidth content as painless as possible, Apple would need the kind of distributed architecture I've been describing at Google. Maybe Apple has that, but if they do, nobody has told me.

There are ways around this technically. Maybe Apple is using the Akamai network or its equivalent. Maybe the distribution system leverages peer-to-peer in the same manner as BitTorrent. Maybe there is some secret technology Apple will pull out from behind Door Number Three. But whatever technical solution Apple presents, the company will still have a LEGAL problem.

No matter how I look at it, this supposed Apple announcement clearly bumps into the claims of little and its nearly 40 video distribution patents. Burst has been quiet lately, which could mean something or nothing. Certainly, they haven't announced that Apple has bought a license like Microsoft did last year for $60 million. If Apple introduces this video service as described but without a license, Burst will undoubtedly try to shut it down.

So if you believe the rumor, then look for an Apple/Burst announcement. I'm not sure yet what I believe.

Comments from the Tribe

Status: [CLOSED] read all comments (0)