About the Laws of Online World Design – Part I

This is the first installment in a series of posts I’ll be making about Raph Koster‘s The Laws of Online World Design, as explained in this introductory post. I will start at the top of the list, and work my way down until I’ve poked and prodded every law in the list, not skipping any unless I really feel like it.

In this, Part I of the series, I’ll concentrate on the following law:

Design Rules

The following law on Design Rules states that the secret of making a successful and lasting online game (goal-oriented, with wide appeal) lies in how your ladders of advancement are setup:

Design Rules

The secrets to a really long-lived, goal-oriented, online game of wide appeal

ideally, make your game not have a sense of running out of significant milestones (try to make your ladder not feel finite)

have multiple paths of advancement (individual features are nice, but making them ladders is better)

make it easy to switch between paths of advancementt (ideally, without having to start over)

make sure the milestones in the path of advancement are clear and visible and significant (having 600 meaningless milestones doesn’t help)

Let’s go through the bullet-points:

Infinite ladders

ideally, make your game not have a sense of running out of significant milestones (try to make your ladder not feel finite)

When taking the first bullet-point to the extreme, your game should ideally have an infinite number of significant milestones which the players can strive for. An example would be a game with no max level, where the players can always become more powerful than they are right now, where there will always be another important quest to complete, always more powerful monsters to slay, always better loot to be had. In other words, ProgressQuest!

ProgressQuest, no matter how basic it may seem, shows that yes, in theory, it is possible to automagically generate a never-ending stream of “content”. However, unless you can create an AI powerful enough to provide unique, meaningful and engaging content for human players – which ProgressQuest doesn’t – on an ongoing basis, you’ll really need human beings of flesh and blood to create your game’s content. This means, that unless you’re the Sultan of Brunei (which you’re very likely not, though I admit there’s a small chance that you may be. If that is actually the case, please forgive me for using your name, don’t sue me and don’t buy my country. Thx.) and can afford to throw money at an unlimited amount of people for an indefinite amount of time, you’ll have a limited amount of manpower to generate new content/milestones in your advancement ladders.

Even if you did had a powerful AI at your disposal which could keep the milestones and the content coming indefinitely, it wouldn’t necessarily make your players stick around for longer periods of time – apart from the small subset of players who would be at the top of your advancement ladders, that is. Unless there’s an artificial roof on those ladders, any fresh blood you introduce into the game world will eventually realize the futility of the insurmountable task that catching up with the veteran players is, and will start looking for greener pastures.

Possible solution: Player-generated content

So what can you do, when you ideally need infinitely long ladders and only have a budget for a limited amount of content? Well, for one thing, you can take a look at what the games that have come before yours have done. Take Ultima Online, for instance. When it was released in 1997, and for a long while afterwards, it had no quests. It had no end-bosses or raid-encounters. It had no linear paths through the game-content from A to Z, which all players followed religiously. At some point it did get a very generic “escort this randomly spawned NPC to a random target town” feature. But what it had in droves, was player-generated content.

Whenever the words “player-generated content” are mentioned, the thoughts of some people (usually the ones working in marketing/legal departments) stray to animated flying penises in Second Life or creatures of a particularly creative nature in Spore, and shivers go down their backs as they contemplate the implications.

When talking about player-generated content in UO, though, it was all about players who lived virtual lives as lumberjacks, tailors, fishermen or thieves. Players who created trade-empires by selling items they had crafted/looted off NPCs (or other players even) on their player-owned vendors which they setup in houses placed at key locations in the game-world. Players who duked it out at the crossroads or player-killers who formed hunting parties to take out unsuspecting players deep in dangerous dungeons – or hunting parties formed to take out the aforementioned player-killers, even. It was about players who created long-lasting in-game communities centered around player-created buildings serving as taverns, magical institutions or which together formed entire towns. Players who took a spoonful of in-game tools (orc-helms) and a shovelful of imagination and took on the roles of orcs, vampires or elves – even though the game didn’t have any support specifically for such game-play.

This happened more than 10 years ago. Obviously the particular game mentioned above has evolved into a different creature since then, but the core remains the same – it’s all about players using the tools offered to them by the developers to generate their own content, set their own goals, their own milestones – every time they log in. And it’s all achievable without having to deal with flying penises, too!

It should be mentioned that MMOG-players of today would possibly be completely at a loss at what to do had you picked them up and put them back down inside the 1997/98-version of Ultima Online. Granted, that was true for many players back then as well, and the reason it worked then could be that the players had fewer expectations and a higher treshold for what they’d be willing to go through in order to keep playing. In most modern MMOGs the players are hand-held from start to finish, are constantly spoon-fed with tips and hints about where to go, what to do next. There are usually no “rails” (officially, anyway) in these modern MMOGs, but players seem to follow them regardless.

While this does offer a more “tailored experience”, it also means that the developers must work hard at constantly expanding the content of the game to avoid players ever reaching the point where they’ve “done it all”. It’s not impossible, though – Guild Wars is one example of an online game with such explicit rails, and as of Feb 2008, more than 5 million copies of Guild Wars and it’s expansion packs have been sold, according to Wikipedia.

Another possible solution: Periodical resets

Another thing worth considering is doing periodical resets of some of your advancement ladders and rankings, similarly to how it’s done in many First-Person Shooters like Counter-Strike or Team Fortress 2. This would level the playing-field every X days/weeks/months and give new players some chance of competing with the veteran players, and would mean the players would always have something to strive for.

In a traditional MMORPG this can obviously not be done to all of your advancement ladders, as the players would stop investing the time it takes to level up, gain equipment etc. if it would all be taken from them anyway, so careful thought would have to be applied when going down this route.

Multiple paths of advancement

have multiple paths of advancement (individual features are nice, but making them ladders is better)

make it easy to switch between paths of advancement (ideally, without having to start over)

The more paths of advancement your game can offer the players, the broader appeal the game will have. The more diverse your ladders are, the more likely it is that any given player will find something to do in the game that they enjoy, and that they will keep doing it.

So what does “multiple paths of advancement” mean, exactly? It could mean several things; choice of which quest-area to level up in, resource harvesting/crafting skills, skill-based advancement systems, individual or guild-based achievements, PvP-ladders/ranking-systems, gear-treadmills, etc.

Many of the above can be found in World of Warcraft. For instance, you could argue that the Alliance and Horde factions in WoW represent two separate paths of advancement. If you dig deeper, you’ll also see that each of those paths have several “sub-paths” of their own; almost all of the character races that make up the Horde and Alliance-factions each have their own starting areas and quests, though those “sub-paths” gradually converge within each faction as the players complete quests, gain levels and move on to new areas. Age of Conan offers a slight variation on this – there, all the characters start out in the exact same area then branch out into the game world at a certain point, before converging to a single point again later on in the game.

A crafting skill can offer a different kind of ladder. In WoW, your character can gradually train her crafting skills as she progresses through the game. The crafting skills have an artificial roof that gradually increases depending on the level of your character, so you can never train your crafting skills to maximum on a low-level character. The result is that you cannot advance on one ladder (crafting) without also advancing on the other (level-treadmill), so if you like the former but hate the latter, you’ll either have to put up with game-play you don’t like or find a different game.

With an achievement-system of similar style to that implemented for Xbox 360-games, various Steam-games and – yes – World of Warcraft, you’ll have alternative paths of advancement which won’t (necessarily) rely on the players staying on leveling-treadmills or playing the “raid for more/better loot”-game, and instead offer new and varied challenges.

In Left4Dead, for example, you can get achievements for doing stuff like… surviving an entire campaign (where one campaign is a number of linearly connected levels) on the hardest difficulty, healing other players X times, killing another player while playing as an “infected” character, etc. You can also offer some kind of reward for the achievements, one example out of many being WoW – where you can gain unique titles, new tabards, pets and mounts by completing certain achievements or series of achievements. When Blizzard released their achievement-system for WoW, it had more than 500 individual achievements that catered to a wide array of players, both hard-core and casual. Systems like this can also be expanded on easily, and can offer a lot of bang per buck spent on development compared to traditional quest/raid-content – though it should probably be a bonus feature rather than the main focus of one’s game, as this is non-repeatable content; once you complete an achievement with one of your characters, it stays “achieved” for that character. That is at least how it’s been handled for the most popular achievement-systems so far.

Then of course you have the multiple advancement paths that PvP can offer. Arena-systems like in WoW can let small, tightly knit groups of players compete against other similar groups in point-based ranking systems. With battlegrounds in WoW or keep-sieges in AoC/DAoC large amounts of players can go head to head and gain PvP-experience points of one sort or the other. Such PvP-XP could also be given to players in “World-PvP”, i.e. random combat between players in an open environment without any of the rules or limitations that are present in the smaller arenas/battlegrounds. You can even take it to another level and let entire factions or nations fight versus eachother in the open game-world, with additional features that let players capture key locations or objects, which could result in a virtually never-ending struggle for power between the player-factions. You also have more generic PvP ladders and rankings, based on kill/death ratios of players, top barbarian of the week, best dueller (don’t forget PvP duels!) of the day, etc.

Switching between paths of advancement

It’s usually not a problem to switch from one path of advancement to another when those paths are separate features as many of the examples described above are. However, when you start talking about one of the main features in almost every MMOG – character classes – you run into trouble quickly. In most of these games you pick a class during character creation, and your character has to stick with that class throughout his virtual life.

Ultima Online didn’t have any classes. It had some basic templates you could choose from when you created your character, but they did not put any artificial limits on what you could do with your character later on. Ultima Online’s main paths of advancement (aside from the player-generated content mentioned earlier in this post) were instead the 50+ skills you could train in. All players could have from 0.0-100.0 (originally, later increased) points in any of the available skills, and you gained points by using the skills actively. No player could max out every skill, they all had a specific limit on the total amount of skill-points they could acquire, and once they reached that limit they would not gain any more total points. What they could do, though, was to re-distribute the skill-points they had already acquired to any of the other available skills. At first this was done through an atrophy-system where the player slowly started losing skill-points in skills that went unused, but later the players could mark any of the skills with an up-arrow, a down-arrow or a “lock”-icon to prioritize which skills they wanted to gain points in, which skills they wanted to lose points in, and which of the skills shouldn’t move neither up or down.

This was a system that offered multiple paths of advancement (50+ independent skills like mining, lumberjacking, arms lore, herding, begging or swordsmanship), as well as made it relatively easy to switch between those paths – without having to start over. Over time, you could go from a miner to a swordsman to a magic-caster with the very same character. Compare that to games like WoW, AoC, WAR, LotRO, etc. Though you can in (all of?) those games switch from one gathering/crafting profession to another for a fee, you will be permanently stuck with the class (warrior, mage, rogue, cleric) you pick at character creation.

Another game that easily allows you to switch between classes – or “jobs” as they’re called in this game – is Final Fantasy XI. There, the initial job a player chooses when creating his character determines his starting-equipment and what abilities he has available to him, but he may at any time (though limited to specific locations in-game) switch between the six basic jobs that are available to the character from day one, and level up each of those jobs independently from the others. If he reaches level 30 in any of the six basic jobs, he gains access to an additional 14 jobs to choose from.

Clear, visible and significant milestones

make sure the milestones in the path of advancement are clear and visible and significant (having 600 meaningless milestones doesn’t help)

All of the below examples represent a significant milestone in a player’s path of advancement, whether it is gaining a new level and getting access to some cool abilities, finally affording that really expensive mount or getting hold of a legendary epic item. They are milestones that the players work towards and look forward to reaching, and will give them a sense of accomplishment when reached. Examples of significant milestones:

Reaching a new “tier” of levels – 10, 20, 30, etc.
Gaining access to new, cool abilities
Gaining access to new crafting recipes
Beating a new or particularly hard “boss”
Completing the hard, final quest in a long quest-series
Getting more powerful equipment
Seeing a new fatality-move for the first time
Buying your first mount, or player-house
Getting to the maximum character level
Maxing out a crafting-skill
Getting a “first kill”-achievement on a raid boss
Getting all achievements within a specific category (like exploring all of Kalimdor’s zones)

The below examples of less significant milestones are just that due to the fact that they do not (in most cases) on their own give the players any feeling of accomplishment. A barbarian in Age of Conan will kill hundreds if not thousands of picts, panthers, spiders and wolves, and killing any single one of them will not result in a whole lot of excitement. Examples of less significant milestones:

Killing a random fozzle
Gaining access to a slightly more powerful version of an ability you already have
Getting a stand-alone achievement (like having hugged four squirrels, or having hopped 15 times)
Gaining a single level, for instance from level 23 to 24
Gaining a fraction of a skill-point (for instance 10.2 to 10.3 fishing skill in Ultima Online)
Completing a single, easy quest in a long quest-series
Being logged in for an hour
Buying yet another cheap accessory/decorative item for your house

In many cases a significant milestone relies on a less significant milestone for it’s significance, or even on other significant milestones. Getting to the maximum character level wouldn’t be a significant milestone if it was the first thing that happened, but because the player has to gain (in most cases) several levels before reaching the maximum, it’ll be a milestone the player gets closer to with each step (level). and a build-up of “momentum” occurs. In other words, without the smaller, insignificant milestones, the significant milestones lose their significance. Which I guess means… all milestones are significant! heh :)

Anyhow, this pretty much sums up what I have to say in Part I of my “Thoughts on The Laws of Online World Design”-series of posts. Feel free to leave comments/feedback/rants/flames! Until next time, I’m out…

Design Rules