Think what you want about Web 2.0, it is an unescapable change in the way the web will grow in the future. It is the convergence of several emerging ideas and technologies that individually wouldn't mean as much but that taken together make a real difference. These ideas are (paraphrasing Wikipedia and isolating the technical points from the business ones):
- "The transition of websites from isolated information silos to sources of content and functionality". This is the "programmable web", which enables a real web or network of applications to share information. It implies the use of open standards such as RSS and XHTML which make the blogosphere bigger than the sum of the blogs.
- "Open communication, decentralization of authority, freedom to share and re-use". This is the community aspect. It consists mainly of user-provided contents (that can range from user comments to fully user-created sites such as Flickr, Myspace or Wikipedia) that the users own.
- "A more organized and categorized content"
- "The resurgence of excitement around the possibilities of innovative web applications"
So what does this all have to do with games? Well, when I started developing software more than 25 years ago, I started developing and playing games (which I still do as much as I can). That's where I'm coming from, it's my culture in a way, so I tend to make a lot of analogies in everything I do with the gaming world. About ten years ago, I launched Bad Mood, an open-source project to reverse-engineer and emulate Doom and build a level editor on the Atari Falcon (don't laugh, it actually worked, and that was way before iD released the source code). The goal of that evolved from simply running Doom's WAD files to extending the format. This part of the project never went any farther than a set of specifications but it's the part I got the most excited about, and I think the ideas in there still stand true today (although I'm not claiming to have invented the concepts). Better, I think they make a lot more sense today because of many converging factors, like Web 2.0 is the convergence of several factors.
Doom was one of the first games to develop a real ecosystem of user-created contents. Ever since, many games have tried to reproduce its success, and games like Half-Life have brought it to the point where users have been creating completely different games from the engine. This is very analogous to the Web 2.0 user content principle. With Bad Mood, we wanted to create an open format of game files that standardizes the description of a game's world. You would have been able to run a RPG, adventure, platformer or FPS game on any engine that understood the format. There is another analogy with the Web: like web sites that have to run on multiple browsers and platforms, games have to run on multiple and widely different platforms like the PC, Xbox, Xbox 360, PS2, Gamecube, Mac, PSP, GBA, DS and even cel phones so a standardized description that abstracts as much as possible makes perfect sense.
Another major idea in Bad Mood was the concept of portals. The principle of portals was that any game or map could contain a set of portals that would initially be closed. Anyone could then claim a portal (which could be free or have a price), open it and connect his own world to it. This looks a little like hypertext where a link can lead to another document with the important difference that the target of the link doesn't pre-exist: the link precedes the target. This is also analogous to wiki contents that can get created after the placeholder was created.
In other words, we would be creating a hypergame. The Multiverse concept is of course very close to that idea.
A side effect is that a game aquires a location as it connects to other existing places. This reminds me of a physical idea that I've always loved, which is that geometry can emerge from a network of relations between objects but that's a different story...
The system could be as open as the Web if it relies on URLs where the contents can be downloaded.
A third piece is necessary for this puzzle to exist, and that is the notion of gamer identity. If you're going to travel in a multiverse, there must be a part of your avatar that persists between games. This is probably the hardest part to define because it can heavily influence the gameplay. For example, you don't want Halo's plasma grenades to travel with you to Kameo's world (well, it would be fun in a way). Xbox Live has the beginning of that concept in that you have a unique gamer tag and score as well as a few universal settings. But how difficult would it be to enable the incredibly detailed character appearance editors of Oblivion or Second Life to be used across games so that your visual identity travels with you?
Finally, the emergence of gaming networks like Xbox Live and MMOGs (massively multiplayer online games such as World of Warcraft) are the glue that binds it all together. Having a gaming network that works across games is obviously enabling those scenarios, and MMOGs bring a large number of gamers in the same space. Marry the two and you have huge potential.
UPDATE: Christopher mentions that Spore, which was demonstrated at the GDC, is a good example of Gaming 2.0. It has almost exclusively user-created contents and it has this interconnection of worlds. It is really an ecosystem in every way. On the other hand, it is still an "intragame" in that it communicates only with itself. These ideas will truly reach their full potential when games start integrating with each other and form what could be called the intergame.
UPDATE 2: Wired had a related piece in its April issue:
So welcome to Gaming 2.0, a world where all games are connected, users create contents and have a well-defined and persistent identity. It is inevitable :)