GDC08: Raph Koster's 'Reinventing MMOs, a Metaplace 'antemortem''


Raph Koster, president of Areae, and Sean Riley, Lead Programmer, held a panel the final day of GDC '08 called 'Reinventing MMOs: a Metaplace 'antemortem'', which was all about dissecting why the concept of MMOs needs updating. Going in, I was hoping for insight, which I got, but wasn't expecting a lot of technical talk, which I also got.

And the takeaway I got from this session is this: Koster's doing some good, forward-thinking work, but Metaplace doesn't really shine for me just yet. Details on both after the break.
I'm going by my notes here, but I'll begin by saying that these are my conclusions based on what I understood of what Koster was saying; I may be mistaken. To start, he gave a few reasons why MMOs need reinvention. First, they're too hard to make. It's true that it takes a large team, many years of development, and a huge budget to try to compete with WoW on their terms. This alone makes Metaplace really important in the current age of user-generated content on the rise. However, I saw a couple of in-development MMOs at GDC this year that are the exact opposite of the 'must beat WoW' mindset -- tiny, independent studios making innovative games that just need to be seen to be successful. Then again, they're a long way away from finished, and they point back to the main theme: the genre itself needs to be infused with something new to grow.

Second, the number of canceled products, the 'get it perfect or go home' phenomenon. How many at-first promising titles have gone away due to the failure of the mass audience to accept them? Think about Fury and its promise of innovation, and how miserably and quickly it tanked. Tabula Rasa itself seems on the fence at times. It could be argued that the lesson to be learned here is 'know your audience', but then where does that leave innovation? If the audience wants it, then innovative titles should flourish. If the audience doesn't want it, then they're doomed to failure.

Third, publisher pressure is at fault for a lot of this as well, as Koster points out. Companies see the success of WoW and think that emulating it will bring profits. But WoW itself has been fine-tuned over the course of its lifespan, and to its credit, continues to mature and refine. It's foolish of any new developer to think they can isolate convenient bullet points to present to publishers to prove a successful model can be built on emulation. Yet that is exactly the kind of presentation that gets titles seed money.

Fourth, an expansion of the innovation idea, is the reinvention of the wheel. Having your game's chat system look and behave just like everyone else's is either laziness or adopting a standard that works. Either way, it's tiresome, and a good indicator that change is necessary.

Fifth, and this is huge for Metaplace, MMOs in general are poorly integrated with the rest of the Internet. Frankly, this is not something that has bothered me -- I've not missed being able to access webpages within Mythos, for example, or chat with AIM friends from Guild Wars. Having said that, however, there is a rapidly-growing community of modders out there who are building tiny applications to enact just this sort of functionality within their favorite MMOs, and there's no reason to think this broad audience effort will slow down in the coming years. It is, in fact, one of the things that Metaplace was conceived to tackle right out of the gate.

So, proceeding from these reasons why MMOs need reinvention, Koster moved on to the following points on how MMOs work today:
  • monolithic giant servers
  • all services contained within the server
  • complex server cluster architectures
  • tight dependency between client and server
He went on to say that there are virtually no successful re-uses of MMO servers, which I'm including here to benefit those who might be interested in this bit. Frankly, when talking about tech, my mind begins to wander. He went on to mention the way things used to work: client (telnet) to server, then back to client, a relatively simple architecture, compared to today: client to multiuser servers, with a static database, a runtime database, a chat server, process manager, authorization system, rmt/market system, armory/metadata, patching ... yoicks.

So after all this complexity, cost, overhead, etc., what's the alternative? Koster goes on to say there exists a system which is:
  • low cost to develop for
  • extremely scalable
  • robust and pretty future-proof
It's called The Web. At this part of the session, Koster gives his reasons why Metaplace is the answer to every problem outlined above. 'Put worlds on the web, not the web in a virtual world', he says, making the point that games are just another type of media on the web. He then proceeded to launch Metaplace live and went over its features. Now, I need to step back and say that I realize that this is still a work in progress, and the idea is not to compete with the graphical prowess of, say, LotRO, but rather to democratize the process of making an MMO and develop a platform that will be a springboard to future successes in the MMO space. Having said that, however, when you show the world what you've been working on and it fails to impress, like it or not, it takes away some of the thunder of your arguments.

Some of the Web-integration Koster demonstrated right away. Appearing in a communal space, his avatar walked around interacting with a bunch of testers, all of whom shared the same avatar as he wore. From the chat field, he typed something in French, and Metaplace instantly translated; typing '/video funny cats' brought up several YouTube selections in an embedded window very quickly, with no loss of connectivity or framerate stutter from the communal area window. The integration is there and it works well.

And then he showed how easy it is to create an MMO using the UI with a few mouseclicks. Unfortunately, what he showed was a room with his avatar walking around in it, sharing space with what was obviously a bot. His point was that it's just that easy to create a place for people to come visit just like any other virtual world, but the secondary point was also made: democracy is a system that allows for the cream to rise at the same rate as the crap. Metaplace is a tool that will allow thousands-to-millions of everyday users to create the MMO/virtual world of their dreams ... but how many of those will be compelling?

Now, I'm not about to say that this isn't worth developing; far from it. We, as a game-playing culture, need this in place to provide a wider audience to all the crazy/awesome ideas for design, gameplay, interactivity that would never be approved by a traditional publishing house. A lot of these ideas will sink; a few will float. Of the ideas that float, a portion will get picked up by development houses and turned into 'real' games, complete with publishing deals and big budgets. A secondary portion will become so popular that in order to remain alive without getting swamped with bandwidth usage fees, the creator will either institute banner ads or ask for donations. A third portion will remain free to play for as long as possible before disappearing, or have their idea co-opted and modified only slightly with a graphical facelift by someone with better artistic skills, before disappearing.

Metaplace will make all of these things possible and pave the way for the truest apotheosis of user-generated content. But what Koster showed at GDC I found utterly uncompelling. Personally, I prefer my YouTube in a separate tab, and my IM sessions kept separate. There is a prevailing attitude that convergence is not only inevitable, but desired and necessary. However, for humans without the mental bandwidth to accept all these competing types of signal simultaneously, it is a dizzying prospect -- a look at a world where you have to run just to stand still. It's the exact opposite of Thoreau's 'Simplify, simplify, simplify', and it's insidious. The message is that if you don't keep up, you will be left behind. This is an argument of quantity over quality, which is a typically American approach.

When Koster finished clicking a few links and declared 'I've just created an MMO', I thought 'Well, no, you've just created a graphical chat room.' I'm sure he was referring to the potential of this new space to become whatever he wanted, but what he was showing was as far removed from a modern MMO as any IM session. It remains to be seen whether Metaplace is of sufficient prowess as to make the kinds of MMO people have in mind -- all of which are sure to be either modeled on WoW, or a reaction against it. People play these big MMOs because of all their features, not in spite of them. If Metaplace can't offer these same features, innovative or not, there's a chance it will be relegated to cottage industry status.

Metaplace's success also depends on the idea that more people are interested in creating their own things, rather than just wanting to play something out of the box, which is much easier, and I'm not entirely convinced that this is the case. Certainly, it's Second Life's main attraction, and how many residents are also content creators? How much of the World Wide Web's webpages have been hand-coded by their users? But then again, does there need to be a large number of creators for Metaplace to succeed?

Again, don't get me wrong, I'm not against Metaplace; in fact, I'm all for it. It's a big step in the right direction, and Raph Koster is absolutely the visionary to spearhead it. He's respected, knowledgable, passionate, and speaks well. Is this the right time for Metaplace? Could be. All I know is that I entered the session excited, and left it unmoved. Surely the potential is there, but if it's up to the worldwide audience to make this platform a serious concern, then it's got an uphill battle ahead of it.

This article was originally published on Massively.