I've been thinking about how long I've been playing this game, and how I was going to post about it--and how I could do so without making myself out to be a fossil. Of course, it's questionable why I don't want to appear a fossil; I get something of a kick out of being an "eminence terrible," and the only real audience I'd mind appearing a fossil to consists of cute geek chicks, and probably not many read my blog.

But anyway.

If I recall correctly, I first played Cosmic Encouter at a Lunacon, sometime in the late Jurassic, at a time when Deborah Harry had not yet become a star--or about at the same time, anyway. I was inveigled into a game by Evan Jones--who I haven't seen in at least a decade, but apparently co-designed the new Marvel Universe RPG with Dan Gelber (co-designer with me of Paranoia).

I continued to play it throughout college, primarily with my college buddies Mark Malamud and Carey Hammer. (Last I heard, Mark was a middle-wig at Microsoft, and Carey founded the interactive department at, if memory serves, Scholastic.) [Geeks all, obviously.]

And when I was director of R&D at West End Games, I published a dreadful version of Cosmic--possibly the worst print edition ever published. Thankfully, Mayfair and, later, the Avalon Hill division of Hasbro, republished better ones. (Though Hasbro has since killed its Avalon Hill division.)

Cosmic is the example par excellence of what Richard Garfield calls "the exceptions game." An exceptions game is a game with very simple basic rules--which are then extended and changed by "exceptions" printed on another component. In the case of Cosmic, each player is dealt an "alien card," printed with a special power that changes the basic rules.

This is a very useful concept because in any game--print or digital--there's a limit on how much you can expect players to be willing to learn before they start to play. If anything, there's more of a problem with the players of digital games, who typically want to plunge in without learning =any=thing first. (And, indeed, this is why, I believe, sophisticated games have grown so radically during my lifetime--with a paper game, a player has to learn not only how to play, but also how to operate the game. With a pre-digital hobby game, you have to learn how to perform combat calculations, add up terrain effects, determine lines of supply, and/or check armor effects versus opponent level--its complicated, and requires a degree of intelligence and commitment. With digital games, you click and go.)

From my perspective, the idea of the exceptions game, along with the idea of "programmed learning"--pioneered in Avalon Hill's original Starship Troopers game, back in the 70s, but followed by every extant realtime strategy game--by which you are introduced to new rules gradually, scenario by scenario--are fundamental to making essentially complicated games accessible to a broad public.

Or to put it another way, Cosmic Encounter is one of a handful of games that are hugely important from an historical perspective; and if you wish to understand the evolution of game design, you =must= understand Cosmic, and its role in the evolution of the field.

And you should therefore play it. At once.

That being said....

I have two severe qualms about this implementation of Cosmic.

First, it absolutely cries out for a tutorial mode. I am an experienced player of the boardgame--but there were at least two occasions when I was at a loss, where it took me a fair bit of time to figure out how to play. The first time when was I wanted to invite other players to ally with me; it wasn't at all clear that I had to click on their icons, rather than their "power" name or their "user" name to do so. The second was when I wanted to make a deal with an opponent, and it wasn't at all clear to me how to offer something in exchange for a deal--I should have clicked on one of my planets, to offer a base there, but there was no indication that this is what I should have done.

Yes, I figured these things out, after looking at the FAQ. But I ain't no dummy, and if =I= had to look at the FAQ, God help Joe. Q. Gamer. Cosmic Online needs a tutorial.

My second problem? It's got to do with the pricing.

When I talk about pricing, I guess I have to give you an idea where I'm coming from. I'm willing to spend a high proportion of my disposable income on games. Back in the day, I was willing to spend hundreds of dollars a month playing games on the old commercial online services.

But the reality of the industry at the moment is that very few games support subscription prices. Sites like Uproar and pogo.com offer classic games for free. Most subscription services---like EverQuest or Dark Age of Camelot--charge $12.95 or so a month, but incur substantial hardware, bandwidth and (not least important) community support costs to do so. I can think of only one game that isn't an MMG that is successful and supports a subscription fee--that's Laser Squad Nemesis. LSN is an exceptionally cool game, and costs $25 for a 6 month subscription.

Cosmic asks $39 for 6 months.

N-no.... I'm sorry.

Unlike MMOGs, this is not a game I'm going to play excessively and exclusively. I can see paying a chunk of money for lifetime play; I can see paying a lesser chunk of money for a year. I can see paying that much for a site where, say, I can play Cosmic, and Diplomacy, and Settlers of Catan, and, oh, I don't know, Bohnanza.

But.... This is a game that lasts 20-30 minutes. And it's a good 20-30 minutes, and I suppose there's a level of persistence with the high-score and "m-rich" system.

But it doesn't seem to me that the pricing is in line with the value received. And that's a shame, in a way; this is a cool game, it ought to be able to find a venue online, and at the monthly price ($8.95), I can certainly see myself, at times saying, fuck it, I'm bored tonight, I'll pay that for 2-3 games.

I'd love to see this game as a free download with several weeks free play and sub thereafter at, say $20 for 6 months, on RealArcade or Shockwave.com (the latter ought to be interested, as the client is pure Flash). And I wonder whether aggregating several games of this type for a single monthly sub might be worthwhile.

Of course, Peter Olotka seems to have "subscribed" me for the next 10,000 years or something, so I really shouldn't complain, and should simply praise it to the skies and get you dweebs to drop your money on the game.

And I do love the game, and wish it all the best. But...

We need a business model that lets this kind of game survive--and thrive.

I'm not sure Peter and crew have found it.
posted by Greg at 12:56 AM

Last month, T-Mobile Germany and CipSoft launched what is probably the world's first massively multiplayer mobile game.

Since I first got involved in mobile games, I've argued that they can come into their potential only through multiplayer gameplay. Mobile phones are relatively primitive computing devices; you can get a better media experience with other machines. Their strengths are ubiquity (you take your phone everywhere) and the fact that they're networked. They're machines used for social purposes--and games with a social, multiplayer component, are what will work best on them.

Unfortunately, the technology providers don't seem to have grokked that. Most phones still only support HTTP; there's virtually no support of TCP sockets or UDP (both better gameplay technologies in most cases), and only a few high-end phones have Bluetooth or another local networking technology for fast-action, local-area gameplay.

And the carriers don't seem to understand this, either; by and large, they are not willing to share airtime or data traffic revenue with multiplayer game providers. They'll share the application download fee--but if you run a multiplayer game, you have continuing costs in terms of hardware and bandwidth to provide the game to people. If a multiplayer game is highly popular, you can wind up =losing= money--your share of the download fee eventually eaten away by your ongoing costs. To make this work, you need a continuing revenue stream to offset your continuing costs--which means you need to persuade your partners to offer a share of the continuing revenues =for them= that your game generates. (I believe this will happen eventually, of course, but the carriers haven't really understood the implications, I believe.)

The result is that the mobile game landscape at the moment consists largely of reskinned arcade classics and puzzle games. Mind you, I =like= both arcade classics and puzzle games, when well done, but I'm a multiplayer gamer at heart.

And a mobile MMG is something of a holy grail--something that would be really cool if only it were possible to build. But you aren't going to do it on a little Java phone with a 64k MIDlet size limit.

So how did T-Mobile and CipSoft pull this off?

Tibia Micro Edition (TibiaME) is based on CipSoft's earler PC MMG, Tibia. (Tibia is free to play, with pay for premium services--there are both English and German versions, so check it out.) The CipSoft crew were basically MUDers who started work on a MUD with a GUI at university, and decided to turn it into a commercial venture when they graduated. They didn't have an Everquest-style budget to play with, so produced a game that's reminiscent of an earlier generation of PC and console titles; Tibia uses sprite graphics atop tiled terrain, in an oblique overhead view, with a somewhat cute graphical style reminiscent of the early Zelda. But it's a genuine MMG, with several character classes, magic, monsters, and all the rest. And hey, it's free--which makes sense. Given the relatively modest technology employed, it's unlikely they could charge the kind of fee that more elaborate games charge--so their model (free to play, but charge for premium services) makes perfect sense, too.

According to Stephan Vogler at CipSoft, they started work on a mobile version of the game early in 2002. They figured that with more recent handsets, they had devices with the power to support a game like Tibia. In other words, the modest technology and graphics used for the Tibia client made TibiaME possible; the Tibia client is a few hundred K, by comparison to the multiple megs required for conventional 3D MMGs. And 2D graphics don't require the same horsepower.

Toward the end of 2002, T-Mobile decided they wanted to move into multiplayer mobile games. They'd launched a GPRS network in Germany, wanted to encourage more data traffic, and games are among the few data services people will willingly pay for. They also wanted to announce a game at CEBIT, which wasn't that far away--and CipSoft was able to demo the game on a phone.

So working rather quickly, CipSoft slapped together a version of the client software for Series 60 phones. Series 60 is a set of extensions to Symbian, the "smartphone" OS supported by most major mobile phone manufacturers; originallly developed by Nokia, they've made Series 60 an open standard, in an effort to encourage a critical installed base of compatible handsets, and some other manufacturers (notably SonyEricsson) have also produced Series 60 phones.

Series 60 was a sensible target platform, because S60 phones typically have processors in excess of 100Mhz in speed, and application memory of several megs. They also allow you to run native Symbian OS applications, programmed in C++, allowing you to dispense with the overhead of the Java (or BREW) virtual machine, meaning the app will run faster. A port of Tibia was clearly technologically feasible. (And T-Mobile had recently announced support for Nokia's Series 60 phones, and was presumably eager to encourage consumers to adopt them as well.)

CipSoft was eager to keep the application size down, however, as they expected to allow users to download and install it over the air. Even over a GPRS network (which offers faster download speeds than vanilla GSM or CDMA), downloads typically take a second per severak kB, and all things being equal, the longer the download, the fewer the users with the patience to complete it. In the event, CipSoft was able to keep the client software down to around 150k, consuming 300 something k of application memory space once installed.

CipSoft had to substantially modify the game to work on a mobile phone--not because of the underlying technology, but because of the UI. A game designed for a windows-and-mouse GUI is not going to work well when controlled from a T9 phone keypad. TibiaME is simplified in some ways (e.g., two character classes instead of 4; all magic through item use rather than 'spellcasting'). As a result, the serverside code was also modified to suit--meaning that Tibia and TibiaME are separate games. You can't play your PC character from your phone, or your phone character from your PC. (I think this is a major flaw, by the way--perhaps an inevitable one, but half the appeal of the concept is the potential of continuing to play your MMG character when away from your desk.)

So--how does the business model work? In a rather interesting fashion, actually. There's no cost to download and install the client; no monthly subscription fee. All the gamer does is pay T-Mobile the normal data transfer cost. (That is, GPRS network operators do not normally charge for data traffic on the basis of "airtime minutes," because GPRS networks do not require a handset to keep an open "channel"--the phone only consumes network resources when sending or receiving data, so it makes sense to charge on a per-byte rather than per-minute basis.) T-Mobile offers several different payment plans; depending on the user's plan, he pays somewhere between 0.09 Euros and 0.018 Euros per 10 kB of traffic. CipSoft says a user will transfer something like 400k over the course of an hour of gameplay, so the ultimate cost is between 3.6 Euros and .72 Euros per hour.

At least at the lower end of the spectrum, this strikes me as very reasonable--you get close to 20 hours of gameplay for roughly the cost of one of the more expensive monthly sub MMGs. (Of course, I was willing to pay $6/hour for gameplay over the old commercial online services, back in the day.)

One issue I can't judge (since I'm not in Germany and don't have an S60 phone) is how gracefully TibiaME deals with the issue of latency. Latency over the air network is on the order of seconds--over the wired Internet, it's on the order of 100 milliseconds. That obviously makes a big difference in any multiplayer game that isn't turn-based. CipSoft claims they have 'prediction' algorithms that mask the effects of latency from players--and player-vs-player combat is disallowed, so latency issues may be less visible. Nonetheless, I've always viewed this issue as the big obstacle to mobile MMGs--and I'd be surprised if TibiaME handles it with perfect smoothness.

T-Mobile reports fairly modest usage so far--but there is a fairly small installed base of Series 60 devices at the moment. Nokia claims more than 10m such devices will ship worldwide over the course of 2003, however, so the user base may grow substantially over time. We'll have to see.

Still and all, this is an interesting development, and something of a landmark for mobile games.

You can find out more about TibiaME in a case study on the Forum Nokia site.
posted by Greg at 11:41 AM

Recently, Bruce Rolston wrote me about an article he was working on for Computer Games magazine. The article was about Edward Castronova's research on economic activities in massively multiplayer games (in which he famously decided that per capita GDP in EverQuest was somewhere between that of Bulgaria and Russia). Rolston's questions were basically a) do game developers take Castronova's idea that the economies of MMGs are "real" seriously; b) if so, does the fact that players are creating "wealth" impose responsibilities on game operators beyond those included in their terms of service; and c) to what degree do people actually "live in" the worlds of MMGs. Some of my responses follow.

Do MMG Players "Create Value?"

I found Castronova's analysis very funny, and interesting. But no, I don't take it too seriously. Here are some reasons why:

1. It's true that it's possible to auction stuff from MMGs on EBay--but very few players avail themselves of the opportunity. In most cases, that's because they treat playing the game as entertainment alone, and don't themselves take seriously that it's economic activity. In most cases, when they leave the game (average duration of a subscription is now around 7 months, and declining), they simply abandon their avatar and its possessions. There are, to be sure, a minority of players who do auction things off for realworld cash--and an even smaller minority of players who purposefully set out to gain virtual items or build valuable avatars for sale. But this is a very small minority. (One of them, incidentally, is Julian Dibbell, author of My Tiny Life, which is well worth reading; Dibbell's blog follows his attempts to monetize his gameplay, and is quite funny.)

In other words, most of the 'value' ascribed to virtual game artifacts will never be realized. The same is true of some real-world endeavors, of course--housework creates value, in a sense, but isn't measured by GDP, and when you vacuum your rug, you don't really think of yourself as participating in the economy. There have been some attempts to quantify this "economic activity," but I don't really see the usefulness of doing so.

2. While there is a market for virtual world artifacts, it's a black market. Virtually all MMGs make this kind of trade grounds for expulsion of the game. And buyers (and sometimes sellers) need to be very wary; there's no mechanism for enforcing trades, which in fact is one reason why the game operators bar asset sales--they don't want to have to deal with customers complaining about the fact that they were defrauded by someone.

It's true that a black market is still a market, of course, and the goods and services it produces are real contributors to economic activity, even if hard to capture in conventional GDP measures.

While it would be possible for a game operator to build its own in-game auction system, with enforced terms of sale, there are two reasons to avoid doing that. First, many game players object to the idea that other people can get better by spending money instead of time in the game--they feel it's unfair (even though most MMGs are not directly competitive--one player's gain is not another player's loss). Second, I'm not sure it would be commercially worthwhile to develop such a system; although most sellers would migrate to a system with enforcement (one area where EBay's network-effect value can be overcome), the service could take, say, 10% of sales at most in service fees. Given the likely volume of sales, I'm not sure it would be worth the cost of building a custom auction system--I did a back-of-the-envelope calculation a while ago, and came to the conclusion that the ROI would be small or negative.

Unless, of course, such a system changed player behavior a lot, so that virtually everyone auctioned items or avatars at times. Which is not inconceivable.

3. The 'value' created in MMGs is valuable only because of artificially-imposed rarity on the part of the game operators. If they wanted, they could create any number of maximum-level characters and any number of hugely powerful items simply by making a few changes in the database. So the "work" that creates "value" in the game is an artificially imposed structure. It's not like the work actually creates anything of real world utility. Nor are game developers interested in arranging their artificial constraints to create maximum economic value of items in the game; they're interested in arranging their artificial constraints to create maximum player satisfaction and retention. Scarcities are imposed for reasons of game balance, not to create valuable virtual artifacts. In the real world, if I can figure out a way to duplicate an item cheaply, I'll make a fortune; in MMGs, dup bugs are the bane of game operators' existence, and will be squashed (and their exploiters banned) as expeditiously as possible.

So it's hard to take the notion that this is 'work' creating real economic 'value'.

Do People "Live" In MMGs?

As to whether people are "living" in this world--it's a matter of semantics, I suppose. Certainly hardcore MMGers spend a lot of time playing--typically 30+ hours per month. (I suppose there may be some who spend 40 hours a =week= in the game, but that strikes me as scarily obsessive.) And for some, real friendships are formed online, and the game becomes a 'hangout,' valued as much for time spent with guild buddies as for the gameplay it offers. (Gordon Walton's famous line: "They come for the game, they stay for the community.") But I wouldn't call MMGs places where people live; they're more analogous to the local bar than a nation. You go there to hang out with your buds and kill a few gnolls (which is at least better from a health perspective than knocking back a half dozen brewskis).

Do MMG Operators Have "Obligations" To Players

As for the notion that because economic activity is (by some people) performed in MMGs that the game operators therefore have obligations to players beyond those laid out in the operator's terms of service--sorry, I don't think so. Firstly, in most cases, the game operators consider that realizing the "economic activity" by exchanging "production" for real world money a violation of the terms of service, and I don't think it's reasonable to say that the existence of this rules-breaking activity creates an obligation to support such activity. Even when a game operator allows asset sales, there's a clear presumption by anyone with a brain that you're joining a game, the servers may go down someday, you have to abide by certain rules (like not hacking the game client) to keep playing, and so on. If an MMG world is a "nation," it's a bizarrely governed one--a private, profit-making subsidiary of a corporation. But all its "citizens" know this going in--and also know that they have complete freedom of emigration to a game with rules they like better, if they object to how this one is run.

The relationship between a game operator and the game's players is often a contentious one, because operators update games, tweaking the rules, continually, and any rule change is going to adversely affect =someone=. Consequently, well-run games often do behave something like a "government," trying to satisfy as many of their customers as possible, provide a safe environment to play in, and trying to keep on the good side of their players. But this is just good business; an MMG's income stream depends on keeping as many people happy for as long as possible. The behavior of game operators isn't dependent on the perception that "virtual economic activity" is happening in the game, but that the game operator's =real= "economic activity" depends on sucking up to the player base.

Should Game Designers Study Economics

As for whether designers of MMGs have economics text books around--in most cases, probably not, but they should. In fact, the same is true for games of all sorts. Games, like the economy, and like the ecosystem, are complicated interactive systems governed ultimately by a single factor: money for the economy, energy for the ecosystem, and the "objective" of the game for games (whether that's winning, character improvement, or some other goal posed for players). Just as the need/desire for money shapes (but does not determine) economic activity, the goal of a game shapes (but does not determine) how people play the game.

For MMGs, public choice theory is particularly important; just as you have to consider the incentives faced by individual members of an institution, you have to consider what any change to a game does to player incentives. As an example, when it first launched, Ultima Online had an elaborate economic system underpinning it, in which all items consisted of some mixture of resources, and resources were constant across the game. So, for example, a rabbit was made of one unit of fur. You could kill a rabbit, and make a fur shirt (one unit of fur), which would eventually decay or be destroyed, and go into the game's "fur pool," ultimately being used to spawn a rabbit. (Or perhaps one half of a wolf--you get the idea.)

The problem is that the way to get better with your tailoring skill is to make a lot of shirts. Consequently, people made far more shirts than anyone actually needed--with the result that there were huge piles of shirts and no rabbits, because any rabbit that might appear would be turned into a shirt, and the game's whole supply of 'fur' was tied up in useless inventory. (UO ultimately abandoned the whole system of limited resources.)

But again--the reason why game designers should know something about economics has nothing to do with the fact that some few of your players may be trying to make actual money off your game. The reason economics is useful is that it gives you a set of tools and ideas for understanding gameplay issues that are analogous to economic issues.

Did Castronova's idea come as novel to people in the games industry? Yes--that is, it was novel that anyone would take seriously the idea that people are creating anything of actual monetary value in the game. But MMG developers (and MUD developers before them) have been worrying about the economic systems of their games basically since their inception.



posted by Greg at 10:45 AM

So Unplugged has launched its first two BREW titles... Skirt Fighter (Verizon and Alltel) and Trophy Fishing (Alltel, and Verizon soon). Woot!
posted by Greg at 11:16 PM

So on appeal, a US district court holds that games are protected as free speech. Well, duh. The original ruling (that games had no idea content and thus weren't "speech") was patently silly. Nice to know the court system isn't totally insane. "This isn't a free country, but you still have that right," as the saying goes.


Gregory Fischbach, co-founder of Acclaim has left the firm. A pity; in an industry whose management landscape often seems depressingly filled with jerks and morons, Fischbach always struck me as a remarkably genuine. But he's getting up there, Acclaim may well be unrescueable, and if it can be rescued, it will take a lot of hard work. Acclaim has always been an odd firm; it was founded as a comics publisher (they bought Valiant, and tried, without much success to rescue some of the old Western Comics titles, including such oddities as Magnus Robot Fighter--incidentally edited for a time by the redoubtable Theresa Nielsen Hayden). They moved into computer games during the brief enthusiasm for "CD-ROM as a medium," and had some success with titles like Turok Dinosaur Hunter.... and had the usual ups and downs of small publishers in a hit-driven industry. In the last few years, it's been mostly down.

Evidently 3DO is looking for a buyer; they declared bankruptcy recently. Another sad story, really; unlike most CEOs faced with a cratering company, Trip did not duck and cover, but instead plowed much of his personal fortune (gained as co-founder of EA) into trying to rescue the firm--a brave and honorable, if (as it turns out) foolhardy thing to do. 3DO is another odd firm--originally founded to launch the 3DO player (I still have mine), the last major attempt (before XBox) to create a US-designed console system. The problem with the venture was that 3DO built a reference platform and got other companies (notably Matsushita) to build it--meaning that the manufacturer got no part of the platform royalty, hence had no incentive to subsidize the price of the hardware. It launched for something like $700--and though a vastly better machine than other consoles on the market at the time, it was just too damn pricey, and died. (Is Nokia paying attention?) Trip recovered by transforming 3DO into a second-tier game publisher--and produced, in the Might & Magic series some games I loved (not, though, the recent crop of releases, which I've previously lampooned as "Coasters of Might & Magic"). I never could understand why they kept flogging Army Men, though.... bleech.

Damn, wish I had a couple of billion lying around... In addition to these two, Vivendi Universal Games, Interplay, Sega, and probably Midway and Titus Interactive are in play... With enough capital, you could rather rapidly put together a company large enough to challenge EA, and large enough to avoid the small publisher trap. (It's a hit-driven industry; if you publish a handful of titles annually, you're chances of generating a major hit are smaller.) Of course, all of these companies are hurting, and it would be a major undertaking to integrate them and turn them around... But I'm astonished there aren't more media companies (and vulture investors) looking at this. There's a huge opportunity to buy into the world's fastest-growing entertainment industry at rock bottom prices.

But then--if I'm so smart, why aren't I rich?
posted by Greg at 10:35 AM

As promised, here is the Powerpoint of my presentation on the game industry's crisis (at least as I see it), given last week at the Digital Genres conference at the University of Chicago.
posted by Greg at 11:23 AM

...but inspired by the fact that actor Hugo Weaving plays both Elrond and Agent Smith.

"You seem to live two lives, Mr. Baggins. In one, you are a peaceful and productive resident of the Shire. In the other, you flit about the world in the company of wizards, dwarves, and other low lives, apparently attempting to hurl a magic ring into a volcano.


"One of these paths has a future, Mr. Baggins."

posted by Greg at 6:41 PM