Spurred by this and a half-dozen other articles, I'd like to explain why I think PC gaming is not dead, dying, or even dogpaddling in the kiddie pool and screaming that it can't swim.
A market will always be there.
As long as PCs are capable of gaming, people will be playing games on them, and other people will be making those games. Desktop PCs might very well disappear in the next 20 years, perhaps to be replaced by a non-centralized interface and AI that follows you from wall to wall to ceiling to countertop to oven to dashboard to [any flat surface you can think of], like a dog you cuddled too much as a puppy and now just won't go away. That is, afterall, the ultimate goal of Microsoft's LiveAnywhere project. When that day comes, there will be games for that platform. But until then, millions of people will play games on their PCs every day... and often when they're supposed to be working.
Modes of profit and delivery can and will adapt.
In fact, they already are adapting. As the first linked article mentions, companies are already experimenting with the use of advertisements to fund games. Some games will integrate advertising seamlessly and invisibly, but don't think the annoyance other games cause will prevent advertising from catching on. The first TV commercial wasn't three frogs croaking "Bud... Weis... Er!" -- it was probably three jerks, obviously high on Mountain Dew, yelling at the screen how Dapper Dan's hair gel could change our lives. Yet, despite our almost-constant loathing, we've come to the point where roughly a third of primetime television is commercials.
If you doubt this development for a second, consider that Battlefield: Heroes (which looks like a fun game, incidentally) will place its advertisements at the loading screen between levels/battles... just like TV commercials between plot events. Hopefully, the additional profit scheme, undisruptive microtransactions for things like appearance, will stave off greed long enough that we won't see 5-minute loading screens for another 10-20 years.
And let's not forget franchising -- the new industry buzzword. Even if Spore is pirated left and right, Maxis will recoup those losses from custom trading cards, creature models, and any number of other peripherals. On top of that, they'll be providing a number of online services which strongly reward connection to official (key-accessed) EA servers, thereby making pirated copies of the game weak and shabby by comparison. A thief hidden in anonymity is much more likely to respond to positive incentives than to threats, because he ain't scared of you.
Far from being on the decline or soon to be replaced by Asian business models, I bet we've only seen the beginning of subscription gaming. Prices and plans will vary greatly, largely because savvy developers will realize that they can start publishing content within six months or less if they simply abandon the nonsense that is the traditional MMORPG world. Start with less, expand more often -- made possible by increased dynamics and decreased emphasis on power and balance. Of course, that's only one of many development styles that will take advantage of subscriptions.
Games will be better
The quality of games matters. The PC platform, like console platforms, will occasionally have a bad series of years on the development side. The original Xbox was the victim of poor development throughout its life. Aside from the Halo games, few other games released on that console were worth a damn. Today, the 360 is the most profitable console out there -- selling more games per console than the Wii --because it hosts so many quality games. In fact, 9 times out of 10 when I hear someone raving about the Wii, they're raving about Wii Sports... which comes with the console (I hope to own a Wii eventually, so don't take this as an attack). When PC game sales slump, a major factor is probably the abscence of quality-focused and reasonably limited game design on that platform.
"Reasonably limited?" I mean games, like WoW, that don't spit on owners of older PCs through insane processing requirements. I mean developers thinking about what they could do with a fraction of the funding and team size that has become the norm, rather than automatically accepting any resources available to them. I don't mean designing the next great solitaire game for Flash, cellphones, and a 5-year-old's Etch-a-Sketch. Games don't have to be quick, easy, or shallow to appeal to millions.
What sort of PC games are selling best? Of the top 40 best-selling PC games of 2007, nearly all are open-ended. MMOs, Sim games, RTS games, Oblivion, S.T.A.L.K.E.R., Diablo (yes, still!) ... almost every one is a game with months or even years of gameplay, most with a strong social component. Even on consoles, what are the most anticipated games of this year? The ones with dynamics and replayability: GTA IV, Fable 2, MarioKart, etc.
open-ended gameplay + simple but smooth graphics and a limited scope = lower costs, greater revenues
But then, I'm not a dev, so why listen to me?
Speaking of open-ended games: