Thursday, March 8, 2012

The Business Model of Neal Stephenson's T'Rain

I picked up Neal Stephenson's latest novel, Reamde, hoping for creative ideas on virtual worlds and business models.  Unfortunately, the book spends way more time than I would have preferred on random modern day gun battles - which are less interesting in the world of 201X than they are in his more famous cyberpunk works where there are more gadgets for the characters to play with. 

The novel features a fictional virtual world called T'Rain - a subscription MMO with a currency that is fully exchangeable into real money, with professional gold farmers/sellers as a core demographic.  Players can subscribe with a credit card, in which case the money from any purchases or sales of gold are settled directly to the card.  There is also an option for "self-sustaining" accounts to pay their subscription fee in gold.  The details aren't fully fleshed out - this is a novel, not a design document.  Still, there are some interesting ideas to be found here.

The Exchange Rate 

The game's lore presents the act of selling gold for real world money as the character sacrificing that gold to a local deity in the hopes of receiving good fortune, while the magical appearance of gold when the player invokes their credit card is similarly presented in-game as divine intervention.  That said, I'm assuming that under the hood is a functioning currency exchange in which every purchase represents a swap of gold and cash between two customers for several reasons:
  • The book spends a lot of time discussing how, rather than having infinitely respawning mobs appear in the world with cash on hand, T'Rain was modeled by geologists.  Gold must be mined by player-owned miners.  (Presumably, monsters obtain gold by killing NPC merchants players pay for stuff?)  This entire discussion becomes moot if the amount of gold in the world isn't constant.
  • If you can both pay your subscription in gold and have the company pay you money for your excess gold, it's possible to have customers that are an actual net loss of money.  While there are theoretical situations in which this might be okay - perhaps if a dozen players subscribe THINKING they can be that one guy who turns a profit, or because they enjoy trying to kill that person in game - the whole quagmire is easy to avoid if the company doesn't gamble its own money.  
  • From the company's perspective, it makes more sense to require players to convert their gold into a constant amount of dollars to pay the subscription fee, as compared to accepting a constant amount of gold and attempting to sell said gold (or generate new gold) to other players.
That said, this approach means a floating exchange rate, which may delay transactions and will, more importantly, impose limits on the amount of money that players can change out in one fell swoop. 

Star Trek Online's Currency Exchange - as of the time of the screenshot, anyone looking to sell much more than $1000 worth of Cryptic points is going to begin shifting the exchange rate downwards. 

Effects of Gold Farming on the Currency Market
It's explicitly stated that a character parked in a "home zone", such as a mine, will be safe and will automatically perform certain "bot-haviors", such as mining gold.  It's also stated that literal Chinese Gold Farmers will have large number of these fully legal bots, all paying their subscriptions with gold naturally, and operating at a net profit. 

Less clear is how much effort/risk this requires on the part of the player.  The implication is that more remote home zone sites are likely to be more profitable mining locations.  The player must somehow safely transport the gold to a money changer to collect their profits, and it is stated that the amount of gold you can purchase for $73 is a lot of money in T'Rain terms. 

The question becomes crucial because of the time - and to a lesser extent risk - curve.  If any player can maintain one or more characters at a net profit (in gold or real money) for limited time and low risk, this will have obvious negative effects on the value of gold (at least until players have picked the world of T'Rain clean of its geologically modeled mineral reserves).  The theory is that with lower cost of living/per capita income, you might have parts of the world where even a small net positive cash flow in dollars would be enough for a gold farmer to live on.  Putting this into practice is a very delicate balance of having it be easy enough for someone to rely on the profession for their real world income but hard enough that the potential purchasers of gold don't just roll up another mining bot of their own instead. 

Subscription Fee As Incentive
The other interesting tidbit is that T'rain's subscription fees are charged on a per-character basis, and are based on character class.  No specific numbers are given, but it is stated that someone intending to play a powerful Warrior-mage is paying much more than it costs for a literal Chinese Gold Farmer to run a character as a mining-bot. This in turn creates some unintended social trends, as players with less real-world money run around as cheaper horse-archers or other more basic classes.

Metaphorically, one could argue that we have similar systems in non-subscription games today, where the conventional wisdom is that the overwhelming majority of players pay nothing and the stereotypical minority overspends to provide most of the revenue. 

As far as the class balance question, I suppose lower subscription fees are amongst the few things we have NOT yet seen anyone try to get people to play less popular roles like tanks or healers.  Perhaps turning the system on its head and just charging people more to be DPS is a good solution.  Or, perhaps it's shortsighted - do you really want players choosing based on price, disliking the game experience, and quitting outright? 

The Fictional Virtual World
At the end of the day, the actual experience of playing T'rain sounds not unlike the notorious space free-for-all of Eve Online.  You have the potential for full looting in PVP, players betraying their factions, skill training based on real world time, and others - my guess is Stephenson drew some inspiration from that world.  Overall, though, I can't give the book any more than a mixed review because of how little time the characters spend in the virtual world that's supposedly the book's focus.  


  1. Ayup, lots of things that I've been thinking about Reamde, said better than I could express them. I get that there's no real difference, computer-wise, from T'Rain Gold buried in a mine somewhere, to T'Rain GP, to $$$ in your bank account. It's a neat plot point, even if Stephenson doesn't explain the system in a way that makes economic sense. I'll suspend disbelief that the T'Rain devs have figured out a working business model using virtual gold mining (maybe lots of Mechanical Turk style front-ends, or something)

    But I'm an (ass-backwards) political theorist, not an economist, and what bugged me about T'Rain was that it used a feudal political system. There's a reason why people dump feudalism about as fast as they can figure out an alternative, and that is that it requires a *ton* of day-to-day work to maintain a feudal relationship between peasants, manorial lords, and royalty. You need constant surveillance and rituals at every level to make sure people aren't shirking their duties in both directions, and because everything is tied up with land and blood, it's very hard to get rid of a lousy person in a feudal system.

    I stay out of MMOs because I don't have the time commitment for anything that even looks like a job in a virtual word. I couldn't begin to imagine the time investment required to manage a whole hierarchy of Chinese gold farmers.

  2. my feelings reading Reamde some months ago were very similar:

    first 300 pages: this is fantastic!
    second 300 pages: this is a long diversion...
    third 300 pages: this diversion is the main plot? seriously? ><
    last 300 pages: Stephenson always writes good conclusions, just go with it.

    As for the business model, it was more persuasive to me than to you: the points that A: resources were allocated in a predictable manner (predictable by geology, that is) means crafters can make rational decisions; B: full death penalties apply, meaning that players need to organise themselves in some manner; C: the game provides mechanisms to facilitate that organisation, both by design (the feudal model) and emergently (the Brightness war); means that you have a functional economy - and yes, i was thinking about EVE as well.

    The point about cashing out was also a good one, because the act of cashing out *destroyed that value in the game*: cashing out in the book was, and in any game would be, the single biggest money-sink ever invented. That alone would probably be enough to mitigate any and all inflation caused by looting dead mobs.

    But yeah, that whole terrorism bollocks? Made the MMO-setting just a MacGuffin, unfortunately :(.

  3. I think he predicted Diablo 3 :)

  4. I'm pretty sure that the online game isn't actually the book's focus.

    It's certainly the hook that draws a lot of the characters together, but it's not what the book is about, specifically.


Comments on posts older than 14 days are moderated and will not appear until manually approved because the overwhelming majority of such comments are spam. Anonymous commenting has unfortunately been disabled due to the sheer volume of comments that are defeating Google's spam filter.