Wired on Friday: How would you like a CD album that self-destructed after you listened to it twice? Examples like that are usually cited to convey the worst effects of digital rights management (DRM).
Media companies will use the software that protects their work from being pirated to deliberately cripple the new digital versions of existing products. According to the pessimists, DRM will restrict even copying that you're currently permitted to do.
You'll never be able to record another TV or radio show without having to pay again. You'll be forbidden to copy and paste even a short extract from a book. Even music playing in the background of your home videos will be seamlessly edited away, unless you're willing to pay an extra performance fee each time you watch your child's school play.
Other computer experts believe that such protections are impossible to enforce. Someone will always find a way to make copies of media that they've already bought: someone will find a way to hack around the restrictions. And others say that we, the consumers, simply won't buy into DRM - we'll stick to paper books, VHS videos and good old-fashioned CDs and DVDs rather than move to a future where we can't permanently own what we buy.
But what if consumers want self-destructing CDs? You won't pay as much for a self-destructing album as you would a normal CD - but what if it only cost you five pence? What if, sandwiched between the price of renting a video and buying a DVD, there's a whole world of other prices and other products you might consider?
This is the optimistic view of DRM: that instead of crippling software and media, it might open up new possibilities - a sliding scale of usage that will let sellers find a price point that will fit every buyer's purse.
Such possibilities might not be restricted to media. As more of our life is controlled by computers, similar contracts might be enforced for other business transactions.
Suppose you purchase a car on a hire-purchase plan. The car's lock could be designed to keep track of your payments. If you fail to make a payment, the car will lock you out and allow the loaning company's bailiffs in. It will repossess itself, in other words. At the end of the loan period, this capability will be deleted and the car will be yours.
That's the vision being pursed by researchers such as Nick Szabo, a computer scientist studying at the George Washington University Law School in Washington DC.
This area of software is known as "smart contracts": the idea that the small print of legally binding contracts might be coded into the very products those contracts are concerned with.
To an extent, this is already happening now. At the simplest end, a standing order is a smart contract of sorts. The computers of your bank regularly transfer an amount from you to your creditors. The contract is fulfilled without human intervention. It might be automatically cancelled, too, if you don't keep up your part of the bargain: the computers at your insurance company, for instance, might suspend your policy if this month's payment does not clear.
There are two main advantages of smart contracts. First, they can be far more complex and flexible. You know the contract you have when you buy a book, largely because most of its clauses are really just physical characteristics of the book. You can't endlessly copy it; you can't sell it to a hundred other people; you're allowed to pass it on to friends, but if you do you can't read it at the same time they do. A smart contract, such as the DRM rights of an ebook, could be far more varied. For 15 cents, you can get a book you can read once. For an extra 75 cents you can read it for a month. If you pay €20 you can give copies to five friends, and so on.
The other advantage of smart contracts is that their enforceability is built-in. If you don't keep up the payments on your car now, the loan company has to pursue you through the courts. With smart contracts, your car just stops working.
As Mr Szabo says, possession stops being nine-tenths of the law. Behaviour becomes nine-tenths of the law. A piece of hardware with a contract built-in becomes more than a legal agreement - it becomes a physical quality of the artifact.
Or does it? Remember that many experts believe that DRM will always end up being hacked. Won't smart contracts - with so much riding on them - be hacked too? Rather than a world where no-one can break a deal, will we have a new black market in those who can re-program smart contracts?
For smart contracts to be accepted, a lot of work needs to be done in the field of computer security. It's not surprising then, that many of the greatest advocates of smart contracts are concentrating their efforts on building more theoretically impregnable computers: mostly by sketching out new operating system designs that are not prone to the system-wide invasions of viruses, worms and malware to which current PCs are vulnerable.
But, like it or not, these vulnerable, hackable machines we have now are the product of their own complex bargains. Security is hard: it costs money, and it introduces a level of inconvenience.
Suppose whenever you ran a new program, you had to agree to everything it might do to your computer? And suppose, rather than having a simple choice of media, or car rental, you had to pick between hundreds of intricate and varied deals? We have our insecure computers because we've chosen cheapness and convenience over slightly more expensive, slightly more cumbersome impregnability. That's the smart contract we've struck with software makers.
It remains to be seen, when faced with a more "secure" system such as DRM or smart contracts, if we end up making the same choice.
Computers may be smart enough to create rich, complex contracts but perhaps - at least outside the heights of the financial sector - we're not smart enough to consider them.
