Here's the game for the day: consider the technology advances of the recent past, those announced for the near future, and whatever you think might be coming down the pipe and then extrapolate to predict what the videogame scene will be like in five years. In particular, what can you do today that you won't be able to do in a few years? I have my own ideas of where this is all headed and from where I'm sitting it looks a little grim.

Remember when you first came home with Quake and typed in your CD key so you could play it? No? Oh, then maybe it was when you popped your shiny new Quake II CD in the drive and had to type in that long code on the back of the game's case. Wait, that's not right either. Ah, now I remember, I was thinking of Quake III Arena. Yes, that had a CD key! Something important happened there, and I bet you barely noticed at the time: a new, somewhat restrictive technology was gradually introduced by the game companies and by the time Q3A came out, it was just part of the standard experience of buying a new game.

Over the years, this trend of adding more burdens has continued, and it will only get worse. Quake players didn't find themselves looking for a no-CD hack and Half-life players didn't need to connect to a master server to play single-player games, but DooM III and Half-life 2 owners just might have to.

In fact, Valve's Steam, and any like services that follow its lead, are the really big step toward a new world of games. Despite the current delay, I still believe that Half-life 2 will launch and that millions of players will buy it within months of its release. At that point, Steam will be distributed to all those millions of computers, building a vast network across which Valve can begin selling software, both for itself and others. Actually, selling isn't quite the right word: they can start renting software. At first, you will be able to purchase outright and actually keep what you buy, but eventually the model will probably evolve into pay-to-play.

This is the model the game industry is evolving toward: one which allows you to access software on the fly, download the content on demand, and pay for every use according to a schedule dicated by the game's owner.

There will be many effects from this transition:

• You will no longer own a physical medium that contains the games you play.
• Parts of those games will exist on your computer at the time those parts are needed, but it is in the interest of the publisher to parcel out, say, a few levels at a time, deleting the older levels as you progress.
• If you no longer own the physical medium, then you will no longer be able to trade the game when you're done with it. So, for example, you can't head down to your local store and get some credit from the used game to put toward buying a new game.
• You won't be able to sell it on auction services like eBay.
• You won't even be able to give it to a friend to try out for a while.
• If there are no games being sold used, then you can't take advantage of the reduced price on used games, as has traditionally been the case.

Also, services like Steam claim to offer anti-cheating features, which when combined with the next generation of Windows security (see below), will possibly hinder the development of modifications of games. I think we may reach the point where it will become a legitimate question to ask "Could Counter-strike have been developed if Half-life had been a download-on-demand game housed within a system that prevents cheating?" Sure, there will be an SDK to allow modification this time, but it will increasingly becomes a question of how much control the game's creators are going to allow for the game player. And, when it comes down to money issues, you can be sure the creators will side with their own interests over those of the player.

There are already examples of all of these processes right now. My understanding is that by playing Half-life through Valve's Steam, your client is only downloading parts of the game as they're needed, caching only the content you're likely to need in the near future. (This may be wrong. I'd be interested in finding a good analysis of what Steam is doing, if anyone knows of such a document.) For years we've had games which existed online only, as downloadable files, and as network connections become more prevalent and robust, you can expect those files to have a "phone home" feature to police their use (i.e. prevent piracy). And there are already games that aren't being resold by the big game retailers: neither Electronics Boutique nor GameStop/Funcoland sell used copies of games like Half-life, Warcraft III, Everquest, or Ultima Online (at least via their online storefronts). The prevalence of the unique identifiers built into these games makes reselling them a liability, since pirates have found ways to get around them. If used games become less common, what will happen to these big retailers, especially since the business of used game sales has been a burgeoning part of their revenue in the past couple of years?

On the issue of software-as-service, there are some precedents worth mentioning. Save the Whales for the Atari 2600 was reportedly only available for download over the Gameline modem service (that eventually went on to become America Online), but was never sold in stores. Until recently, it existed only as an apocryphal piece of Atari history, since no one had actually ever owned a physical copy of the game. Now that a copy has been found, the world realizes that it probably should have stayed lost. Still a piece of history had nearly vanished because software-as-service lives only as long as the service.

This kind of software also lives only as long as you can pay. For example, the academic world is struggling with how to deal with electronic subscriptions to journals. With a paper subscription, you get to keep those you've bought, but with electronic-only access when your subscription goes, everything goes. The same is true of software, which is good for the game company and bad for you.

Other factors are conspiring to make the so-called Wintel PC into a closed box, a box for which user upgrades are discouraged. Already the Windows Activiation required for Windows XP uses a HWID computed from the hardware configuration in the computer at install. Changing too much hardware may require a reactivation of your Windows XP operating system, providing at least one barrier to user upgrades. The next tightening of the thumb screws will come in the form of Next-Generation Secure Computing, which will provide a locked-down environment from hardware power-on to the desktop. Once the hardware itself is involved in the security process, the ease with which users can change or upgrade hardware may again be further discouraged.

What kind of box is that? We normally call it a game console, and Microsoft and Sony and Nintendo already sell them.

And it is interesting to note that consoles are, themselves, headed toward the same sort of software model, selling services instead of a physical product. Each current console producer is experimenting, albeit at different levels of intensity, with online games. Microsoft's Xbox is way ahead of the pack, with copious local storage via a built-in hard drive, and a built-in ethernet adaptor. Both are used extensively for several games, offering extra levels, voice chat, and game invitations, all available only via their Xbox Live! integrated service. Perhaps most importantly, Microsoft has begun offering what they term "premium content" which is available for download, but only at a price.

This is the next big step, selling software to console owners through an online service, that will be most closely watched by Microsoft, Sony, and Nintendo as well as the big software houses like Electronic Arts. If enough people are willing to pay to download an add-on to a current game, they may also be willing to pay to download a whole game itself.

As an example, Electronic Arts could sell a game like Madden NFL Football once in the store, and then charge an annual subscription fee to download an updated engine, an upgraded set of models, and of course the new year's roster. By charging $35 for the update and saving on distribution and media production, they'll no doubt make more money on resubscribers than on the original sale. As a bonus, they can advertise it as a price break for loyal customers, since it won't carry the heavy $50 price tag of the game on physical media. Once the model is proven to work, and it can work, the whole industry will begin to shift that direction.

At that point, the problems listed above for PC games will begin to apply console games as well. No longer will you be able to rent the latest games from Blockbuster to try them out before deciding on whether to buy. No longer will you be able to sell used games when you're done with them. No longer will you be able to lend a game to a friend without having them pay the same kind of fees you paid to activate the game itself, if that is even possible. Moreover, by building network authentication into a game's boot sequence, a company like EA can begin to deactivate games like Madden past a certain date, in effect forcing the users to upgrade, even if they don't want to play online. Since local storage and broadband network connections are virtually guaranteed parts of the next generation of consoles, the ability to sell software as a service this way will become not only feasible, but attractive to the software companies.

Many will say that these things can't come to pass, that the public will rebel at some point. They'll dig up the DivX debacle or other technological boondoggles from the past decade as evidence. Yet, it will only take one best-selling game, like Half-life 2, to introduce the masses to new and more restrictive technologies that will then become standard.

And that's where were headed, like it or not. No physical media. No rentals. No used games. No sharing games among friends. Limited hardware upgrades. Pay-to-play. Unless something seriously changes the course of the industry, this is the future.