Steve Ballmer of Microsoft fame recently said that "we" are in need of personal computers (PCs) that cost $100. Questions abound about that statement: Where should those 100 bucks be spent? What bits of Ballmer's suggested technology do we really need? And who are we required to kill before or after such a purchase event?
I suggest that "we" don't really need or even want a hundred-dollar PC that's of the traditional kind. Instead, a better way forward is applying new-style Digital Fidelity (Di-Fi) computing technology to the problem spaces traditionally occupied by PCs.
If you've ever tried connecting your digital camera, printer, and flash RAM keychain to your external USB hub for picture print and storage, you're a frontrunner in Digital Fidelity computing behavior. Like old-fashioned high-fidelity (Hi-Fi) systems consisting of radio tuners, amplifiers, and turntables, Di-Fi is about lots of hardware pieces that you can mix and match as you gowithout opening your PC's case. Where Hi-Fi was all analog, however, Di-Fi is all digital.
You might be surprised to learn that there's more value overall in such a piece-wise architecture than in a super-cheap shrink-wrapped PC "with everything."
Why PCs Are So Expensive
Overall, traditional PCs are very confusing in terms of price. The cheapest PC I know of is the Xbox, which hovers around $250 retail. It's a PC in a fancy case, with Microsoft Windows inside and a couple of hand controls. If you subtract the cost of the Windows software, you'd be lucky to have $1 left for the hardware. Clearly, PC parts are very cheap for manufacturers, but not so for the general public. Purchase a PC either ready-built or in pieces, and you'll have little change out of $1,000. And yet, PC parts are supposedly commodity items. Something is clearly wrong.
Once it's purchased, the value you derive from a traditional PC is not so high, either. If you use it to write your Harry Potter knockoff, perhaps that's different, but as a piece of consumer electronics, PCs provide fairly poor value. They're supposed to be "instant on" and "always on," but I've yet to experience that convenience (at least, not the way Star Trek portrays it). The PC you get will sound like a truck and glow like a radiator. It will pump out hot air like there's no tomorrow. The obscure and black art of building quiet PCsa task not for the timidis of no help for mere humans, and expensive laptops provide the only real escape route.
Why is it that Microsoft and friends can assemble very cheap PCs and no one else can? It's not as though any of the parts are especially unique. It's not as though PC parts are made in Switzerland, where labor costs are scary and confusing. In fact, there are two reasons for the high cost of today's PC: For the first reason, welcome to the horrid world of container-shipped distribution networks. The second reason has to do with distorted free markets.
Markup and Shipping
Suppose you're a retailer and want to sell an Asian-made keyboard in the West for $10. Here are your business realities. It could cost $50 to fly the keyboard in with a courier, so that won't work. You'll have to do it the cheap way: Work with distributors and pack full a 20-foot-long seagoing container. That's 7,500 keyboards per container, or else pay more to share the space. It's a significant purchase commitment.
If you're accustomed to bickering over prices at swap meets or flea markets, you might be shocked to learn how the supply chain takes the risk out of such large orders. In the country of origin, the remote distributor might take a 60% markup to process the goods. In the destination country, the local distributor might do the same.
For a fast education in the financial witch's brew that is shipping logistics, read this page on the Indonesia Bali Export Products site. I promise, you'll find it interesting.
As a callow youth working in toy stores and department stores, and as a harried adult doing occasional business in the gift industry, I can report that if a retailer discounts small items below a further 100% markup, then it's good weather for snowboarding in Hell. Total markup for a sadly ordinary case involving two distributors and a retailer: (1.6 × 1.6 × 2) - 1) = 4.12, or 412% markup. Finally, there are the shipping costs: an additional $120 per cubic meter (11 cubic feet) of goods if you're moderately well-organized. Try a web site like this one to compare prices. If you measure your keyboard for container volume and apply a range of shipping costs (go ahead), you'll find that up to 25% of that $10 keyboard purchase is due to shipping costs alone.
All of this processing is very exploitative. The shipping company gets only 12 cents. The manufacturer gets $1.40 at the outside. The consumer gets a piece of equipment whose factory-door value might be only 50 cents. There's not much in it for anyone. And it's not much of a business shipping a box of almost nothing around the world. It sounds pointless, doesn't it? No wonder keyboards are despised by our stall-holder friends at computer swap meets. And we don't even get them for 50 cents.
Using the same analysis, it transpires that up to 40% of the retail cost of a CRT monitor is lost to shipping. No wonder manufacturers are enthusiastic about modern thin-screen monitors. CRTs are big and chunky; thin screens pack together very wellcheaper to ship. Worst of all, a PC case can have 75% of its retail value tied up in shipping costs. That's not much money left for the case itself.
So the first problem with traditional PCs is cost-to-market. A prebuilt Xbox saves shipping, but is a closed proposition. A custom-built and extensible box is price-inflated, because of the expensive tour that the pieces took around the world on their way to you. In neither case is the consumer or the manufacturer particularly happy, because a great deal of the retail price went to the middlemen (Microsoft and distributors).
Retail price is a cold way to assess value, too. If I buy a $50 PC case, I'd rather it be one where less than $37.50 (75%) of the $50 was spent wrecking the environment with diesel shipping. I wanted a PC case, not a mortal sin. If I buy an Xbox for $250, I'd like to think that there's at least $50 of actual componentry inside. Call me a hopeful pessimist.
Why We Put Up with It
The royal "we" know why consumers put up with such waste: We're occasionally brilliant, but more usually confused. But why do manufacturers put up with such waste? It doesn't sound like the value-adding litany of modern business to me. Why do they churn out keyboards full of nothing, worth nothing?
The reason can only be that there is a sense of paralysis. The existing PC architecture is not something that a hardware manufacturer can afford to mess with very much. If integrators, distributors, or operating system vendors drop a manufacturer's hardware from the compatibility list, it's a business disaster. It's easier to put up with shipping costs, and have your business survive. Windows is the primary integration point for most PC hardware, and you have to fit in or perish. Once you finish trading licensing agreements with chip makers, patent holders, and proprietary standards consortia, you might still have a compatibility audit with Microsoft to get through. The 50 cents your keyboard is worth is probably nearly gone.
Shipping costs and complex business environments are not good for consumers. They're not good for hardware manufacturers, either. There's no solution while Windows sits in the middle like a fat toad, as the sole integration point.
And that's where Di-Fi comes in.