The future of virtual worlds?
Feb. 5th, 2007 05:47 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
mirrorshardSparked off by a pub conversation with ![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-userinfo.gif)
hairyears and purplecthulhu the other week.
Second Life, frequently hyped as the Precursor of the Snow Crash-style Multiverse, isn't it, despite having quite a lot of flying cars. It's sloooooow, tedious, and full of script geeks and giant flying furry penises. They hype their membership numbers, by allowing them to be reported as the total number of Residents - which counts everyone who's ever signed up for a name there, whether they play regularly, play once a month, played once a month ago, or never finished downloading the client. And their little alt, too. This is more or less the latest (at the time of writing) roundup of the Second Life user numbers fishy business that I know of.
Areae is the new idea from Raph Koster, who designed Star Wars Galaxies (the original system, not the new version), EverQuest, and a number of other things. He's the author of Theory of Fun, which is a great read if a bit cartoony - it's written for game designers, who are instinctively anti-academic. Information about Areae is thin on the ground as yet, but I've been thinking of it as MySecondHotelQuest2.0 - it's likely to have social networking features like MySpace, customizability and personalization like Habbo Hotel, a similar sort of focus on user-generated/user-mediated content as Second Life does, and competitive game-like features to EverQuest.
Oh, and it's going to be very 2.0.
Which is a fancy way of saying it's going to take advantage of the current best practice in what have been regarded in the past as a lot of different areas and demographics, and try to synthesize them all. I don't think it's going to be a problem for them, because there is a lot of space left in this market - to paraphrase Raph, everything that's been done before is a moon shot project, and they've stuck to tried and tested formulae with a few extensions.
The main strands have been, more or less, Diku Muds --> EverQuest --> World of Warcraft and
LambdaMoo --> Second Life. For the first, game-like, set of worlds, the focus has been on graphics, combative gameplay, and things that help that out. (There's also a certain amount of focus on IC pointless bling, which power-players love so long as it is both pointless and not unhelpful, but that's a discussion for another time.) The social networking features are primitive in the extreme, but they exist... in the kind of way we had on the Web five to ten years ago.
There are two salient points about in-game social features.
First, they only work in two dimensions, possibly one and a half. You get to join a guild, one guild, and you get one social channel with all those people. It has a few internal signifiers for rank, and depending on the game it might have some democratic mechanisms. You can also use a friends list to keep an eye on miscellaneous other people, but it rarely supports anything useful like tagging or annotation. You can hardly ever join, let alone create, cross-cutting groups. It's fundamentally feudal.
Second, they're all in-game, and all focused on characters. A lot of gamers - possibly most serious gamers, I don't know the stats offhand - extend that out of the game, and use the web to add the communication and social networking features they want. Usually forums and personal profile pages.
So that's two obvious wins for the next generation to implement, right there. A third is
getting players more involved with the game, helping them feel like they own parts of it, and making their contributions part of the game itself.
The biggest problem with that, of course, is Sturgeon's law. Most of the people who will be wanting to contribute to your game are teenage gamer dudes who know what they like, and they like big dangerous things with lots of loot. Oh, and hot chicks in impractical costumes. They like those too. I'm stereotyping of course, but as always on the internet you'll end up with some really, really good stuff and a lot of meaningless drivel. Peer-editing and Slashdot or Digg style grading/rating can help the good stuff float to the top, but only if you have people with taste leading it, and just as importantly developing a culture of taste overall.
Second Life has done very well at encouraging general creativity, involving both artistic and programming skills, but almost all I saw wandering around there was rather tacky club gear and bling. Part of that might be the frontier problem - there's always more cheap land. If you don't like what someone's done then go West, young man, and do it better. Resources not priced highly enough for optimal utilization, and all that. So there's no social or economic pressure against suboptimal usage, either in the sense of huge empty buildings or of shelves full of similar tack to everyone else's.
One way to start solving this problem is to provide an API, rather than letting people script ab initio. Allowing them to express their creativity combinatorially, as well as through free text and descriptive flourishes, might be a bit limiting for them, but it will end up with less utter shite. And, quite importantly, fewer self-replicating items and even worse exploits.
Let's look at the technical side of game design now. Games designers and theorists have been making excited noises about native scripting languages and runtime interpretation for awhile now, and it would make writing new content for games a lot easier. Currently, as far as I understand it - not having seen any of it myself - it's all monolithic one-off C code running as close to the metal as it possibly can. Which gets the best performance out of your servers. Adding in the extra layers to support that sort of scripting would quite likely degrade things a lot, and these applications are, obviously, incredibly resource-intensive already what with the hundreds of thousands of people running around and killing things in the gorgeously drawn landscapes they demand these days.
One of the reasons Web 2.0 took off like it did was the technical advances that made it possible - better servers, better networking, many more good connections, and the ability to shuffle off tasks onto the client computer with AJAX. Part of it, of course, really is the out-of-control philosophy (cf. Kevin Kelly. To summarize - "let's do stuff and give people tools and see what happens") and the amazing things people do given the chance.
I'm really not sure that exists yet for the game side of things, given the sheer amount of data that needs to be slung around the place, and the security implications of letting the client handle any of it. (Of course, that'll be remedied if someone does manage to implement the extra layers, compartmentalized modularity, and so on. You'll just be able to send the non-critical bits like puzzle games or drinking contests over to the client for computation.)
On the other hand, Raph knows what he's doing, and we can pretty much guarantee that this is going to be interesting. It may or may not be the Future.
hairyears and purplecthulhu the other week.Second Life, frequently hyped as the Precursor of the Snow Crash-style Multiverse, isn't it, despite having quite a lot of flying cars. It's sloooooow, tedious, and full of script geeks and giant flying furry penises. They hype their membership numbers, by allowing them to be reported as the total number of Residents - which counts everyone who's ever signed up for a name there, whether they play regularly, play once a month, played once a month ago, or never finished downloading the client. And their little alt, too. This is more or less the latest (at the time of writing) roundup of the Second Life user numbers fishy business that I know of.
Areae is the new idea from Raph Koster, who designed Star Wars Galaxies (the original system, not the new version), EverQuest, and a number of other things. He's the author of Theory of Fun, which is a great read if a bit cartoony - it's written for game designers, who are instinctively anti-academic. Information about Areae is thin on the ground as yet, but I've been thinking of it as MySecondHotelQuest2.0 - it's likely to have social networking features like MySpace, customizability and personalization like Habbo Hotel, a similar sort of focus on user-generated/user-mediated content as Second Life does, and competitive game-like features to EverQuest.
Oh, and it's going to be very 2.0.
Which is a fancy way of saying it's going to take advantage of the current best practice in what have been regarded in the past as a lot of different areas and demographics, and try to synthesize them all. I don't think it's going to be a problem for them, because there is a lot of space left in this market - to paraphrase Raph, everything that's been done before is a moon shot project, and they've stuck to tried and tested formulae with a few extensions.
The main strands have been, more or less, Diku Muds --> EverQuest --> World of Warcraft and
LambdaMoo --> Second Life. For the first, game-like, set of worlds, the focus has been on graphics, combative gameplay, and things that help that out. (There's also a certain amount of focus on IC pointless bling, which power-players love so long as it is both pointless and not unhelpful, but that's a discussion for another time.) The social networking features are primitive in the extreme, but they exist... in the kind of way we had on the Web five to ten years ago.
There are two salient points about in-game social features.
First, they only work in two dimensions, possibly one and a half. You get to join a guild, one guild, and you get one social channel with all those people. It has a few internal signifiers for rank, and depending on the game it might have some democratic mechanisms. You can also use a friends list to keep an eye on miscellaneous other people, but it rarely supports anything useful like tagging or annotation. You can hardly ever join, let alone create, cross-cutting groups. It's fundamentally feudal.
Second, they're all in-game, and all focused on characters. A lot of gamers - possibly most serious gamers, I don't know the stats offhand - extend that out of the game, and use the web to add the communication and social networking features they want. Usually forums and personal profile pages.
So that's two obvious wins for the next generation to implement, right there. A third is
getting players more involved with the game, helping them feel like they own parts of it, and making their contributions part of the game itself.
The biggest problem with that, of course, is Sturgeon's law. Most of the people who will be wanting to contribute to your game are teenage gamer dudes who know what they like, and they like big dangerous things with lots of loot. Oh, and hot chicks in impractical costumes. They like those too. I'm stereotyping of course, but as always on the internet you'll end up with some really, really good stuff and a lot of meaningless drivel. Peer-editing and Slashdot or Digg style grading/rating can help the good stuff float to the top, but only if you have people with taste leading it, and just as importantly developing a culture of taste overall.
Second Life has done very well at encouraging general creativity, involving both artistic and programming skills, but almost all I saw wandering around there was rather tacky club gear and bling. Part of that might be the frontier problem - there's always more cheap land. If you don't like what someone's done then go West, young man, and do it better. Resources not priced highly enough for optimal utilization, and all that. So there's no social or economic pressure against suboptimal usage, either in the sense of huge empty buildings or of shelves full of similar tack to everyone else's.
One way to start solving this problem is to provide an API, rather than letting people script ab initio. Allowing them to express their creativity combinatorially, as well as through free text and descriptive flourishes, might be a bit limiting for them, but it will end up with less utter shite. And, quite importantly, fewer self-replicating items and even worse exploits.
Let's look at the technical side of game design now. Games designers and theorists have been making excited noises about native scripting languages and runtime interpretation for awhile now, and it would make writing new content for games a lot easier. Currently, as far as I understand it - not having seen any of it myself - it's all monolithic one-off C code running as close to the metal as it possibly can. Which gets the best performance out of your servers. Adding in the extra layers to support that sort of scripting would quite likely degrade things a lot, and these applications are, obviously, incredibly resource-intensive already what with the hundreds of thousands of people running around and killing things in the gorgeously drawn landscapes they demand these days.
One of the reasons Web 2.0 took off like it did was the technical advances that made it possible - better servers, better networking, many more good connections, and the ability to shuffle off tasks onto the client computer with AJAX. Part of it, of course, really is the out-of-control philosophy (cf. Kevin Kelly. To summarize - "let's do stuff and give people tools and see what happens") and the amazing things people do given the chance.
I'm really not sure that exists yet for the game side of things, given the sheer amount of data that needs to be slung around the place, and the security implications of letting the client handle any of it. (Of course, that'll be remedied if someone does manage to implement the extra layers, compartmentalized modularity, and so on. You'll just be able to send the non-critical bits like puzzle games or drinking contests over to the client for computation.)
On the other hand, Raph knows what he's doing, and we can pretty much guarantee that this is going to be interesting. It may or may not be the Future.
