Thursday, October 13, 2005

SWG: Dependencies Among the Professions

[2008/05/16 note: Well, obviously all of this was made completely moot by the Sony/LucasArts decision to rip the professions out of SWG entirely and replace them with nine "iconic" classes. I'm including this essay here anyway as it still has something useful to say about designing a large-scale MMORPG in such a way that players are encouraged to interact with each other.]

An insight I had a while back was that while the original design of SWG established dependencies between the playstyle-based profession groups, these dependencies were all constructed to support combat. I'd like to explore that theory a little more here to see what others think.


In the Elder Days when SWG launched, it was designed so that each group of professions (see the addendum at the bottom of this post) had a valuable role to play. The goal (I believe) was to construct the skills system so that everybody would have something worthwhile to contribute to other players.

The thing was, only one playstyle really needed anybody else: combat players.

Combat players required crafters to create weapons and armor. Combat players required socializers (healers and entertainers) to heal wounds, cure poisons/diseases, supply buffs, and remove Battle Fatigue. As for explorers, combat players didn't exactly require them, but some of their skills were still useful for a combat player to have.

So being required by combatants wasn't a Bad Thing. It meant that each of these non-combat professions had a useful role to play in SWG. Not only did the combat professions need them, they needed the combat professions in order to fulfill their designed support roles.

The thing was, no other dependencies were created. Combatants needed these non-combat professions, but the non-combat professions didn't need each other. Each non-combat profession came to be thought of as useful (and received new content) only to the degree that it supported combat, rather than how well it supported all playstyles. There were some exceptions -- Politicians (once player cities were introduced) definitely needed Architects. But these were rare exceptions; by far the more common case was that combatants -- and only combatants -- needed non-combat professions.

For some time, this wasn't an issue. As long as combatants needed the non-combat professions, everyone had something useful to do.
But -- surprise, surprise -- combatants didn't like being the only ones who had to depend on everybody else. They didn't like having to pay a lot of credits for high-end gear. They didn't like having to hunt for uber buffs. And they definitely didn't like Battle Fatigue. And thus were spawned the buffbots and AFK macrotainer alts.

This was clearly not a desirable state of affairs. Everyone was complaining. And so it appeared that the solution was to reduce the dependence of combatants on other professions. Over time and many changes, the non-combat skills that supported combat play were circumvented, nerfed, or removed. High-end loot drops and quest rewards were added to replace crafted goods. Buffs were greatly weakened. Battle Fatigue was eliminated completely.

The result? Because virtually the entire design of each non-combat profession had been based on direct support of combat play, reducing the dependency of combatants on these professions reduced the value of playing the non-combat professions. Because these professions were never designed to also need each other, and because they were rarely if ever given new content intended to make them fun to play in and of themselves (as combat has consistently been given), removing their combat support abilities leaves non-combat players asking themselves, "What's left? Why should I keep playing SWG?"

Which is where we are now.


Having laid out this theory, I'd like to ask some questions.

1. Do you buy it? Is it mostly right? If there are flaws in the facts or reasoning, what do you think those flaws are? If it misses some other crucial point, what point is that? Or do you think the whole thing is completely bogus and all concerns expressed utterly unfounded? Why?

2. Assuming you see some truth in the theory, do you think anything can be done about it at this point? Is SWG too far gone down the road of all-combat, all-the-time for any attention to non-combat playstyles to save it? Or is there still hope?

3. Assuming you think there's still hope, what do you think can/should be done? Is the answer to create new dependencies among all the profession groups? What should these dependencies be? For example, how should entertainers depend on healers or crafters or explorers? How should explorers depend on combatants? How should healers depend on entertainers? How would you explain to players why these new constraints on their preferred playstyle are a Good Thing?

4. Would an alternate approach -- adding significant non-combat support content to each profession group -- be enough? In other words, can each profession be made so much fun on its own that it attracts players regardless of whether it supports other playstyles or not? Would it be OK to take this kind of standalone approach to a "massively multiplayer" game, or would something important be lost in not providing content that fosters interaction among the playstyles?

As an addendum, here's how I group the various professions.

In SWG and other MMOGs, four playstyles seem to be most common: Combat, Commerce, Exploration, and Social.

Some of these styles can be broken down a little further. For example, in SWG I think of the Bounty Hunter and Smuggler professions (and eventually the revamped Ranger profession) as "Rogue" professions, and of the Rogue group as a subtype of the Combat playstyle. The Social and Commercial playstyles break down to subtypes, too, but the Exploration playstyle isn't detailed enough with professions to have any substructure.

All of my observations above are based on this concept of organization for SWG's professions. If you have any comments or questions on this structure, those are welcome, too.

  • Standard Combat
  • Melee
  • Brawler
  • Fencer
  • Pikeman
  • Swordsman
  • Teras Kasi
  • Ranged
  • Carbineer
  • Commando
  • Marksman
  • Pistoleer
  • Rifleman
  • Space
  • Pilot, Imperial
  • Pilot, Privateer
  • Pilot, Rebel
  • Special
  • Jedi
  • Squad Leader
  • Special Combat
  • Combat Medic
  • Creature Handler
  • Rogue
  • Bounty Hunter
  • Smuggler
  • Crafting
  • Architect
  • Armorsmith
  • Artisan
  • Bio-Engineer
  • Chef
  • Droid Engineer
  • Shipwright
  • Tailor
  • Weaponsmith
  • Sales
  • Merchant
  • Ranger (moving to Rogue)
  • Scout
  • Entertainment
  • Dancer
  • Entertainer
  • Musician
  • Healing
  • Doctor
  • Medic
  • Special Services
  • Image Designer
  • Politician

Wednesday, October 12, 2005

Key Features of the Original Sim City

As an exercise in design, I once sat down and tried to list what I thought were the key features of Sim City.

It's sort of amazing how few key features there really are!

    • Terrain
      • Dirt
      • Forest
    • Zoning
      • Residential
      • Commercial
      • Industrial
    • Services
      • Transportation
        • Road
        • Rail
      • Power line
    • Facilities
      • Active
        • Power Plant
          • Coal Power
          • Nuclear Power
        • Airport
        • Seaport
        • Fire
        • Police
      • Passive
        • Park
        • Stadium
    • Block query
      • Land/zone/service/facility type
      • Density
      • Value
      • Crime
      • Pollution
      • Growth
    • City evaluation
      • Ratings
      • Statistics
      • Problems
    • Taxation
    • Place Block (on dirt terrain)
      • Zoning (large block)
      • Services (small block)
      • Facilities (several sizes)
    • Bulldoze existing block to dirt
    • Set tax rate
    • Player-set
      • Tax Rate
    • Internal
      • General
        • Population
        • Funds
        • City value
        • City score
        • Mayoral approval rating
      • Per Block
        • Population Density
        • Pollution Level
        • Crime Rate
        • Land Value
        • Housing Cost
        • Traffic
        • Police Influence
        • Fire Protection
        • Satisfaction (aggregate)
    • Each tick of the clock is one month.
    • Zones must be connected to power to attract sims; connections are checked each month.
    • Taxes are assessed on all sims at the end of each year to generate funds.
    • Placing a block costs an amount of money that depends on the type of block.
    • Sims enter and leave the city each year based on satisfaction with city-wide and local variables.
    • Connection to services and proximity to facilities will increase or decrease sim satisfaction.
    • Increasing the number of sims in each zone type affects the desirability of other zone types.

Tuesday, October 11, 2005

Inflation and Mudflation in MMOGs

The first thing to notice about faucet/drain economies is that there are actually three things going on in them. There's wealth (in the two forms of items and money) that enters the game through the "faucet" of mission payouts and currency loot drops; there's wealth exiting the economy through the "drain" (item destruction and things like taxes and service fees); and there's wealth circulating in the game among players (people exchanging money and items).

What's important to see here is that only the faucet and drain matter when considering whether inflation is occurring inside the game world. Money and goods circulating among players do not contribute to inflation -- in fact, the more of this that happens, the better for everyone in the game it is.

To understand how the faucets and drains determine inflation, we first need to agree on what "inflation" really is. According to economists, inflation is the condition that occurs when the price of a broad array of standard goods rises over time relative to the perceived usefulness of those goods. Note that term "broad array" -- it means you can't just look at the price of swords only, or at the prices of very rare items, to know whether inflation is occurring. You have to consider the average price of several different kinds of readily available goods. If that average price goes up meaningfully over an extended period of time, then you've got inflation, but not otherwise.

And just to make life more entertaining, there's not just one kind of inflation -- there are at least three.

Standard inflation is the kind most of us think of; this is where a bunch of money enters an economy while the number and quality of goods produced remain relatively constant. In standard inflation, the value of an individual unit of currency decreases over time for most available goods. If 10 dollars today is worth half of what it was yesterday, then an item whose absolute value was generally agreed to be 100 dollars yesterday will cost you 200 dollars today.

Demand-pull inflation is the next type. Suppose you have a fixed amount of money circulating in your economy. Now, slowly cut back on the kinds and numbers of goods being created in that economy, or add a lot of new people to the economy without also increasing production. Over time, prices will generally increase as people consider goods to be worth more (i.e., as demand increases because supply is not keeping up with purchasing power). As prices rise for the same goods over time, an individual unit of currency is worth less and less... and that's demand-pull inflation. (This is something that can happen in a MMOG if crafters are widely supplanted by loot drops for high-end objects, or if you irritate your crafters so thoroughly that they quit your game and aren't replaced by new crafters.)

Finally, cost-push inflation is what you get when the costs to produce goods rise generally. This kind of price increase is usually caused by things like increased wage costs (as through "minimum wage" increases or hikes in corporate taxation) that are passed on to consumers. In games that don't support corporations or that don't have corporate taxes applied by the system, this type of inflation generally doesn't happen. But it can happen if crafting requires natural resources, and the developers cut back sharply on the amount or quality of those resources.

The most common type of inflation in MMOGs is standard inflation. It shows up when the amount of money being created in the game by players (doing whatever the game allows them to do to make money -- usually running quests) exceeds the amount of money exiting the economy in the form of taxes and fees. This can happen if taxes and fees aren't set high enough to match the amount of money players are creating.

This can also happen when there's a currency dupe exploit. If when these happen they aren't corrected by changing the code (to stop the exploit) and removing the money very quickly (so that innocent players don't exchange valuable goods for "dirty" money), a game economy can inflate badly, possibly to the point of ruining the game. So tools for tracking the creation, circulation, and destruction of money in the economy are crucial.

Finally, not every MMOG winds up dealing with inflation. Another potential problem for MMOG economies is deflation (sometimes called "mudflation"). This occurs when valuable objects enter the game world and never leave while the money supply remains relatively constant.

Mudflation tends to happen in particular as developers create high-level content. If powerful items can be obtained more than once and/or can be transferred to other players, then over time the price of low-level or average items of the same type will decline as more of the high-end items enter the general economy and trickle down to younger characters.

Note that a major secondary effect of mudflation is to make many quests and mobs irrelevant. When everyone can afford to buy very good items, there's no need to loot mobs or do quests that yield less valuable items. For a developer, this decreases the value of time spent developing low- to mid-level quest and mob content because now users are able to complete this content more easily than expected.

Various efforts have been made to combat mudflation. The concept of "soulbinding" -- setting the "no-drop" and "no-trade" flags on items -- is only partly to counter twinking; its primary purpose is to prevent valuable items from entering the general economy. Decay and damage effects also help reduce mudflation, though not as effectively as soulbinding.

The main reason that both inflation and deflation are bad news for a MMOG is because they alter the difficulty balance of the game, especially for new players. Because new players have less money than established players, the value of their money is significantly greater. So if new players are unable to buy standard items because of inflation, the starting game will feel too hard. If they are able to buy more items and more powerful items than the designers intended for them to be able to own (due to deflation), their starting gameplay experience can be too easy.

In both cases, these first impressions of a game can be the difference between a long-term subscriber and someone who goes elsewhere to find a game that isn't too easy or too hard.

Designing systems to effectively monitor and manage the economy is non-optional for a large gameworld.