Unhelpful

I’ve written an article about MRPG design for developers. Hopefully players will find it interesting as well. Some of it challenges basic assumptions about how MRPGs should be designed. Some of it details specific ideas that become more and more feasible as bandwidth and CPU power increase.

Origin, Verant, Turbine, as well as the many companies developing MRPGs currently in production: this one’s for you.

Thinking Bigger: Rate = Distance over Time

[Included in its entirety in the extension for your convenience.]

Thinking Bigger: Rate = Distance over Time
Issues in Online Multi-Player Game Development
An Essay for Developers by Delusion

This article represents the design goals I believe would create a new and satisfying step forward in the genre of multi-player online role-playing games. Direct examples from Ultima Online, Asheron’s Call, Everquest, and other games may be helpful in illustrating specific examples.

Although I write this with the game developer in mind, it may be valuable for players as well. Too often, I think we as gamers get used to the status quo and leave it to the game developers to ask the big questions, like “What’s next?”. If we don’t want the next generation of MRPGs to be about cybersex, eBay, credit cards, camping, and laser-toting elves in leather bikinis, we’d best be prepared to figure out what we do want them to be about, and to do that, we have to address you, the developers, before you’ve made your most basic decisions about your game.

Scale

A playing environment in this sort of game can, in my opinion, probably not be too big. Ultima Online, a couple of years after release, feels positively cramped, even on a new server. Asheron’s Call features a world that is vast in comparison. Even so, Asheron’s Call will no doubt feel smaller a year from now than it currently does, and honestly, it’s shrinking fast as the number of portal-casting item mages increase.

There’s more to size than “length times height”. Teleportation plays a major role in how “big” a world seems. UO offers instantaneous travel to anyone who really wants it, whether by direct casting of the recall or gate spells, or through the purchase of recall scrolls to use on a collection of purchased runes. To get from a popular city on one end of the map to another popular city instantly is trivial. In AC, the flexibility of instant transportation is greatly reduced (especially for lower level and melee characters who do not or never intend to add item magic to their skillset). The ability to gate to a specific location requires more of a time investment in AC than in UO - if you need any proof of this, compare the bribes and rewards offered by other players for gating someone from Arwic to Fort Tethana as opposed to a gate from Britain to Vesper, as well as the frequency of the asking.

In addition to spell-based teleportation, it’s built into the game environment itself. In UO, moongates stand outside each of the major cities like bus terminals for those who can’t cast recall, as well as certain other instant teleportation locales. In AC, there is a network of one- and two-way portals. There are few locations that are far out of reach of a portal, so with a little searching around on various player web sites, you can find tools that make short trips out of the most lengthy overland runs.

What if a game had no means of instant teleportation?

The most obvious effect is that the perceived size of the play environment would seem much larger. There are advantages to this; more so if the play environment IS actually bigger. Making a “distributed world” is possible, too: rather than have several identical servers, each server would be different and players could travel between them, though the travel should be difficult. This may not be practical at all, if players tend to congregate in one area of your world, and leave other areas empty. Detail to the proper distribution of adventuring, socializing, and encounter locales might be able to offset this “clumping” effect.

Economy

“Player-run economy” and “virtual economy” are often listed as design goals in multi-player online games. Economy without locality is hard to achieve. Locality is one of the reasons players create their own spaces for trading - whether that be a cluster of player-run vendors in a player-run town outside a city in UO, or a weekly swap meet in an Allegiance Hall outside a city in AC. For other players, locality isn’t necessarily a goal, but a side effect. Whatever the motivations of players, locality and economy stimulate one another.

An instructive example taken from history is the town market concept. In medieval Europe (and in some cities, before then), many cities would set a ‘market day’ for people to peddle their goods. Some sleepy English villages grew into major cities thanks to being early adopters of this concept. Market day still exists to a certain degree, I certainly saw it in England during my time there: it was important to the community. The positive community effects of this arrangement are nowhere near as strong now as they used to be, as market day competes with the economy of scale when a shopping plaza or mega-store opens up nearby. Although the benefits of community versus convenience is a contemporary issue, it’s not exclusively so. Regardless of how we feel this plays out in real life, I think it’s pretty clear that community is good for the stability and longevity of online games, whether that be community out of the game in the “extended play-space” of player web sites and guilds, or a group of players who get used to seeing one another in-game. In the realm of gaming, the “mega-store” mode of economy (such as buying from anonymous NPC vendors who are always stocked with the same items) is less beneficial to the game itself than a player-driven economy: there’s less value added when Jim’s character gets used to buying from Jane the NPC than when he gets used to haggling a price with Fred’s character, and the two start trading with each other regularly, each adding value to the other’s experience.

I call this “local economy”. It’s an important concept that most online games haven’t embraced. The game design itself can lend to local economy. Though none of the current games really attempts a local economy, there’s a slight difference between UO and AC that is instructive.

In UO, game currency gold pieces are rather constant. Characters store them in a bank, and those funds are available to them from any other bank. Some NPCs selling large-ticket items withdraw directly from characters’ banks. To be in Britain and buy a house from a merchant in Vesper, a player simply ensures his or her character has the funds in that character’s bank account (which of course follows that character wherever it goes in the game), then travels to Vesper - be it on foot or through teleportation - and makes the purchase there. Players suffer no real inconvenience or loss by travelling to Vesper and making their purchases than you would if they had purchased it anywhere else.

In AC, game currency pyreals are still constant. Unlike UO, though, characters have no banks. Instead, they have a secondary currency: trade notes. If someone wants a 10,000 pyreal trade note (which represents a sum too heavy to be burdened with for any length of time), the character pays 11,000 pyreals to an NPC. The 1,000 difference is a money-changing fee, and it feels to many players like a service tax, which it is. If the character dies, about half of the pyreals carried in hand are lost, but trade notes are never lost. So instead of UO’s secure bank, AC has something closer to a secure wallet. When the character sells that 10,000 pyreal note, the original NPC who sold it will purchase it back for 10,000 - any other NPC will give 8,000. Instead of a 9% loss, a 27% loss. The effect this has on characters is somewhat subtle, but it confers a definite financial advantage to having a place to call “home” in the game. This is especially true for those without item magic teleportation. I feel the effects would be more pronounced on the player-base if there were more big-ticket items to purchase in AC, but even without it, it does encourage a small degree of local economy.

If unlimited storage banks have a negative effect on local economy, unlimited storage banks that a character can access from any city, such as is the model of UO banks, is even worse. If you’re going to allow players to have what amounts to a safe deposit box, there should be a game currency cost for this service per month, and if it’s not a fixed amount of storage, the price should increase as the weight increases, and should be based on the highest weight amount held per month. For instance, it might cost 3 silver shillings per month if the maximum weight of the box in the month was 5 pounds, but putting in a 100 pound suit of armor in the box (in this case, more like a locker), even for the briefest of moments, would raise the cost to 75 silver shillings for that month.

For currency banks, access to funds should be limited to the bank in which funds were deposited. Most banks will probably limit currency deposit to the currency of that realm: currency from other lands will be accepted at the normal exchange rate, minus a fee. Some border and port towns may allow different currency deposits or waive the fee, reflecting a more cosmopolitan economy. A modest interest, perhaps 1% to 5% per month, may provide a decent trade-off for this extra layer of interaction. To avoid money transfer exploits, this interest should be calculated on the minimum balance for the month. With this additional source of money, it becomes important to ensure that there are plenty of good reasons to spend it. A minimum balance of 10,000 units, kept for 12 months and never withdrawn generates 1,157 units in revenue at the end of the 12th months at a 1% rate, and generates 7,103 units in revenue at the end of the 12th month at a 5% rate. Rates could even be different (and possibly even fluctuate) in different banks to reflect the strength of different nations and local economies.

What if a game had no means of instant teleportation? What if each major city had its own currency?

The game engine itself would encourage local economy. Computer-controlled monsters and NPCs might have a mix of raw metals (and not just gold), and local currencies with an emphasis on the currency dominant in the nearest civilized lands. There would be fluctuating exchange rates based on the influx of materials into a given kingdom. The exchange rates would hit hardest those who had the most money and freedom to travel: successful player characters. An enterprising character might be willing to give you 240 gold Thujan crowns for your 2,000 copper plugs from the next kingdom over instead of the 216 that the NPC-controlled Commodity Exchange offers. What happens when the local silver mine dries up? The currency of the Free City of Harborstone might start using a lower fineness of silver for its silver doubloons, making them worth less in exchange. This complicates matters with the neighboring Principality of Bara, since Harborstone doubloons and those of Bara used to be created similarly and exchanged on par. The Prince might loose patience with the situation, and finally have a means of convincing the Mercantile Guild to support incorporation of the Free City into the Principality, by force if necessary.

Trade

With local economy, and certain areas producing different raw materials and trade goods than others, trade inevitably follows. Trade between nations, trade between cities, trade between guilds, etc. - some controlled by the game political engine and NPCs, some controlled by player characters. In a world without instant transportation, how do goods get from the buyer to the seller? They travel by land and sea.

This implies merchant fleets and caravans. By merchant fleets, I don’t mean a boat like in UO where one person can stuff five tons of metal armor on his one-man 30 foot sailboat and get it from point Trinsic to Ocllo himself. By caravans, I don’t mean a 200 strength runner like in AC carrying 4 complete suits of Amuli armor from Glenden Wood to his guild-mates in Cragstone.

Since our ideal world is a big place where travel doesn’t occur with a wiggle of the nose, let’s assume a land-based caravan going from one city to another. These two cities are at opposite ends of a medium-sized nation, and perhaps that’s a journey that covers one twentieth (1/20) of the distance on the map, and it’s either a straight east-west or north-south line. For comparison purposes, one twentieth of the map in a straight line is half the width of UO’s Britain, or in AC, the distance between Nanto and Mayoi. Our ideal world is much larger. Since these two cities are at opposite ends of a medium sized nation, let’s assume that’s a distance roughly equal to the distance between AC’s Holtburg and Qalaba’r. I can’t easily put that into UO terms, because that distance would be the equivalent of running from the southern tip of the main continent to Vesper many times over. In AC, as a single character with decent run, a flat out run without teleporting would take up two hours or so, I believe. A caravan doesn’t “run”.

What happens on this caravan from Bylayve to Atsetal as it travels the lands of the Kingdom of Estova? What attracts various sorts of players?

A merchant character might be looking to earn a profit by purchasing steel from the mountain town of Bylayve and selling it to the shipwrights in the harbor city of Atsetal, and then purchase glassworks there for the return trip.

A guild captain might be looking to get a large shipment of weapons and armor purchased from the smiths of Bylayve to replace the current guild finery, and from Atsetal, board one of the guild ships for the final leg of the journey home, where his followers are eagerly waiting for replacement of the sub-standard chain mail they’d been using before the guild’s newfound success.

The caravan consists of ten horse-drawn carts. Four carry the steel shipment of the merchant. Two carry the finished arms and armor of the guild captain. Other carts are owned by other player characters and NPCs who only have one cart of goods, or who share part of a cartload with others.. Several of the people on the journey are riding their own horses - in fact, the merchant has hired a bored NPC to ride an extra horse he picked up in Bylayve: he wants to swap out horses for his steel carts, since they’re a little heavier than a single horse can carry for that length of time.

Three of the other carts are owned by NPC representatives of the Mercantile Guild of Estova, and they have two representatives along on the journey. These two NPCs and the merchant and guild captain discuss safety before leaving, and decide to hire a squadron of mercenaries as a security detail for the trip. The mercenaries need to be paid, and if any bandits (or worse) are encountered, their standard contract gives them rights to plunder looted from these adversaries after damages have been paid to anyone losing part of their shipments. For their trouble, they’ll take 3 gold Estova sovereigns per day, rounded up. There are eight of them, and the journey is expected to last four hours. They round up, so the caravan leader will need to pay them 24 gold sovereigns at Atsetal. The caravan leader has dealt with the mercenary leader before, and they trust each other, so the caravan leader pays him 16 in advance. This is a contract between players, and not a part of the game’s interface, as PCs are involved on both sides.

What the PCs involved in the caravan aren’t aware of is that the mercenaries will actually be needed. There’s a hostile nomadic tribe heading through the area that the caravan doesn’t know about, and a straggling group of them will cross the caravan’s path. They don’t like the King, and they blame his people for the murder of their chief. Fortunately, the hunting party the caravan encounters number only three, and one falls to his death during the brief skirmish. When the caravan stops for water, two of the mercenaries need to make camp for the day as their players’ log out - they’ll get half-pay from the mercenary leader the next time he sees them, that reduces the caravan leader’s cost by 8 sovereigns, which means the 16 already paid is payment in full. Of course, the caravan’s risk is increased on the rest of the trip.

Obviously, an online role-playing game needs more than merchants and heroes. What other sorts are served by a caravan?

• Travelers from one area to another who are not able to travel the route alone, have non-combatants, or who are travelling in a much more dangerous area than they’re used to: everyone travels.
• Confidence men and nomads. Those accustomed to living on the periphery of society, gypsy-types, shady businessmen who need to leave town before word gets out.
• Assassins, road gangs, opportunists. Some might try to infiltrate the caravan, others might want to ambush it, try to overrun it when it’s vulnerable at night, and others might have a specific target in mind. Of course, some may be crazed killers, but more about justice later.
• Monsters. Whatever your definition of ‘monster’, an attack on a caravan at 3:1 odds might be attractive. Intelligent civilized monsters might be attracted to wealth, and simple lumbering beasts might pass up a caravan completely if it doesn’t smell of food.
• NPCs - nothing attracts a crowd like a crowd. NPCs would have the same sets of motivations as characters would to join a caravan.

Time

If the current crop of online role-playing games has taught us anything about time, it’s that trying to compete with Earth time is a waste of time. Players don’t use the in-game calendar to schedule events: it’s unwieldy, it generally has no connection to real time, and there’s no way to keep track of it out of game.

Whatever is done, time needs to follow an Earth standard, regardless of what names you call the months and days.

In a truly large world, the time of day should not be a universal constant. One side of a spherical world should be light, and the other should be day. If tied to a 24-hour day, this poses an interesting dilemma to players: If they usually play at the same time of day, they might choose to live in a part of the world where it’s always ‘day’ during their play time. Or if they’re villains, or “wanted” in their country, they might be best served by living where it’s night during their playtime.

That’s poses some unusual considerations. A better solution might be to have a 48 hour or a 12 hour daylight cycle in-game. As has been said before, tying to “real” time is necessary: if you don’t do it as a developer, your players will do it for you. If you have a 48 hour daylight schedule, you may have to come up with an in-game reason the populace of the fantasy world keep track of time in increments other than rotations on its axis. The standard list of fantasy reasons come to mind, such as a planet that slowed down to half its rotation speed, evil magic, a disruption in space/time; keeping to a more “practical” reason might be more useful. Other than the daylight cycle, the yearly cycle, and the lunar cycle, human time is completely arbitrary and based on custom (the concept of ‘hours’ is related to Catholic monastic schedules of the Dark Ages, for instance) rather than laws of physics or mathematics. The more extreme faction of the French Revolution proposed a “day” of 10 daylight hours and 10 nighttime hours, each divided into 100 metric minutes. I only use this as an example of the fungibility of time: if you have a 48 hour daylight schedule, keep track in 24 hour days and throw the discrepancy to historical and social custom rather than crazy magic or weird planetary motion.

What is “Offline”?

Proper treatment of offline time takes care of many issues: travel, trade skills, and sleep not the least of them.

A player log in his or her character, plays for a long session, say eight hours on a rainy weekend. When the player is logged in, the player’s in control in control. Then you player logs off. What happens to the character?

It’s certainly not reasonable to assume that the character just stands (or sleeps) there until the next time the player’s ready to play. Surely the character had better things to do than to stand next to a bed at the inn during the six days the player went on vacation.

What if players could utilize the offline time and assign tasks? Given a proper progressive skill system, you could make offline time more valuable than the macroing which otherwise might take place.

A player’s character should be able to do things when the player isn’t minding him or her*. For one thing, the character must sleep. That’s merely the most obvious thing to do after eight hours of heart-pounding combat, especially for those swinging a 50 pound two-handed steel blade.

What about those characters engaged in trade skills? “Playing” a blacksmith character in a role-playing game seems like an absurdity. Crafting a chain mail surcoat is an exacting task that requires many many hours of work, a high level of skill, and is a job, not a hobby. Creating ten plate mail leggings in 20 minutes stretches credulity beyond repair. More to the point, blacksmithy, bowyery, carpentry and the like aren’t “entertainment”, they’re things characters do for a living - what’s more, they shouldn’t be means of “instant gratification”.

Deciding how characters spend offline time would be more realistic, and could be used to replace a lot of the multiple-click tedium of current online role-playing game trade skills. Clicking 1000 times to make 10 plate chests in 1 hour is neither realistic nor entertaining. Instead, when a player logs off, the player should be able to choose how the character should spend that time. This can either be a simple selection, or preferably a customizable scheduler.

Choices as to how characters spend offline time should be diverse.

• Weapons training. Eight hours of actual combat should remain much more effective than eight hours of offline training with a sword master, but given the choice between having a character do nothing and having a character sweat it out in a training barracks, the choice should be obvious.
• Stats training. In a stats-based game, it might be appropriate to allow a temporary (or permanent in some cases) boost to statistics for offline time dedicated to self improvement. A period of offline time devoted to practicing public speaking might raise a “personality” stat by ten points for an eight hour period.
• Travel. In a huge world, this is an excellent way to make travel more equitable. Allow characters to choose where they wish to log in, and base the radius of their eligible choices by the amount of time they’ve been logged off. In order to lessen “flee exploits”, allow any character who has been offline for more than four hours to choose where he or she logs in. A character who has been offline for five hours should have a smaller radius of login choices than a character who has been offline for 24 hours. A character that’s been engaging in weapons training should have a smaller login radius than a character that’s been practicing fishing, since you can fish intermittently as you travel. A character that’s been engaging in blacksmithy may have to “give up” some of the offline time for his skill if he wishes to instantly travel when he logs in, since blacksmithy keeps a practitioner in a specific workplace. In order not to diminish the value of overland trade travel, characters specifying a new login point should be required to travel relatively light on their way there or deal with very limiting radius reductions if weighted down.
• Trade skills. Any such trade skill where high-quality crafted items or bulk consumables are created by an artisan is applicable. Allow offline time to be non-consecutive: If a full suit of plate armor takes 80 hours to construct, players should be able to devote time to it as they see fit, and finishing once they’ve accumulated 80 hours. If they want to create a true signature piece and have the skills and supplies to do so, maybe another eight hours might be required to add enameling, engraving, or a more difficult design. Likewise, allow a carpenter to make more chairs or axe handles in 16 hours than in eight.
• Spell research and Language training. More on this later.
• Exploration. Again, more on this later.

You don’t want to encourage offline benefits to the extent that being logged in is less profitable (either in terms of monetary gain or character building) than staying logged out, but you do want to make the benefits greater than eight hours of macroing. This can be done by offering offline options (such as trade skills) that can only be utilized in this offline time.

You also want to cap this activity, scaling it so that benefits don’t accumulate after a certain amount of time, such as 48 hours. Requiring two hours of login time before re-setting offline use would be good, too, to reduce the practice of “refreshing” on behalf of friends. Travel time should probably be the only offline ability to accumulate after this arbitrary time period: someone gone for a few weeks should probably be able to log in at any civilized area they could otherwise physically get to when they next log in.

A good customizable scheduler would be at the heart of this system. Players should be able to direct their characters’ offline time as they choose, and distribute it as they choose, assigning different priorities. Players will want to develop a system of priorities to make the best use of their character’s offline time, using a graphically based interface, some of which might be superficially similar to UO’s skill lock interface. For example:

Frank won’t be able to play for three days for whatever reason. He wants his character Alyan to accomplish some tasks while offline. Alyan is a would-be knight, and is currently a squire. Frank sets Alyan’s first priority: practice training with a horseman’s flail until his skill has improved five points. Alyan’s second priority is to travel from the stable in Windheim to the stable in Cheltham, and lodge his horse there. Alyan’s third priority is to complete a bronze sword he’s working on: he has four hours to go. It won’t be a great sword given Alyan’s low ability in this field, but it will cost him 100 reales less to make one than to buy one just as good. After that, Alyan’s last priority is to hunt small game and sell the hides, which isn’t great money, but brings in a few dozen reales. The first tasks are expected to take up about 16 hours of offline time, and the last one will continue until the offline max is reached. Of course, if Frank logs in earlier than expected, the offline schedule is interrupted, but may be continued upon logoff. Frank may either continue with the existing schedule, change it, or make a new one entirely.

* The remaining question would be “when the character is logged off and doing things offline, does that character appear in the world?”. While remaining in the world might be an interesting challenge, players should have control over the appearance and demeanor of their characters. With that as a design goal, leaving the PC to wander around (even if invulnerable) would be an undesirable loss of player control. A player’s friends shouldn’t have to guess if a player’s character is “manned” or not.

Spell Research

With a highly diverse spell selection, researching magic becomes a possibility. UO had no spell research program, EQ had the beginnings of one, and AC went a little further in this regard, adding random reagents to the process.

A big world would be complimented by a broad range of magic, and research as a “doorway” to these spells is a valid factor that serves to regulate the distribution of high power magic. A magic research system like AC’s serves to delay the acquisition of spells rather than truly regulate their distribution. A spell system in a current online role-playing game should be wide enough to incorporate several different types of magic. Many (if not all) magic systems should be dependent on research.

With offline spell research for all but the more basic spells, you can have a great deal of control over how spells are allocated, and more interestingly, dynamically add new ones without having to plug in complex reagent formulae which merely serve to delay the inevitable. You can add a powerful new spell, and specify the level/skill required to properly research it. You can further specify that only a finite number of people per server can learn it per week, or limit it to those researching it in a particular nation, or characters of a particular religion or ethic belief.

The player should have options on offline spell research, such as choosing a certain class of magic (cleric prayers, wizard spells, druid-style ceremonial magic, voodoo-like fetish magic, whatever), and perhaps “directing” the nature of the research. Directing research toward “offense spells” should make discovering new magic less likely, but increase the odds of getting an offensive combat spell considerably. The more narrowly defined the spell type (cleric prayer defensive healing magic, or conjurer ceremony creature summoning, for example), the more focused the research should be - less likely to get magic, but more likely to get exactly what it is the player is looking for if magic is found.

Magic can be made more interesting than it has usually been. While there is certainly a large place in a magic system for reagent/level spells cast by an individual character, other methods of spellcraft are often ignored. If there is any other alternative in existing games, it’s usually clerical magic, which is assumed to be granted by a deity without involving physical components like reagents. Other options are available, too, and they can be added in addition rather than instead of traditional spell magic: group magic crafted by characters pooling their powers, “mental magic” - psionics - that are assumed to come from a person’s inner will, ritual magic like voodoo and witchcraft that involve fetish magic, incantation, and appeasing dark powers, demonology (but only if demons are something truly to fear, and if the chance of having slight control over one is worth the risk of the horrible consequences), etc.

Exploration

Many real-time and turn-based strategy do not allow the player to see the entire playfield. The player can see the areas his or her units are in well, and can see general details about areas where his or her units have passed but are no longer near. This is usually called “fog of war”. A similar system would serve a large-area game well. Although this sort of information will be gathered together and put on fan web sites, having a more accurate map in-game has definite advantages.

The more time you spend in an area, the more accurate your in-game map of that area would become. A cartography or land lore skill could add further benefits. Scrying magic should also serve to improve the map. As a certain area became more and more familiar, the more details you would see: topography, rivers, cities and outposts with their names, encampment areas, dungeons, etc. Players should be able to “compare maps” in a specific fashion, so that one player could trade a certain section of their map for a certain section of someone else’s. The accuracy of this copying could depend on a cartography skill and the level of detail of the source map: it shouldn’t be a give-away. Not only would scrying serve to “flesh out” areas of the map, scrying spells for spying on others should be more effective in areas where the character’s map is more complete. Characters should be able to annotate their maps with location information and other text labels as well.

Offline time could be spent on exploration. Since there wouldn’t be any danger in exploring this way, progress should be much, much slower and measured compared to the amount of detail gained by actual online time spent in those same environments.

Language

One of the context-shattering elements of current online role-playing games (the US servers, anyway) is that the majority of the characters are speaking the same language. The only alternative to this is to either speak in another language, or to make an English pseudo-dialect up, such as the Shadowclan and the Picts have in UO. While this is an interesting solution, it has serious limitations: an “orc” that only speaks “orcish” is going to see the English of other characters and the player has to be vigilant if his “orc” doesn’t understand the common language.

Let’s look an example of another option:

In order to keep this simple, let’s discuss it in the context of a historical role-playing game rather than a fantasy role-playing game. Assume you’re playing a game that sets the PCs as Roman soldiers and their enemies. Your band of Roman scouts are out in the northern periphery. They encounter an angry (but not yet combative) pack of barbarians. Actually, they’re Gauls, but as a typical Roman, you don’t make these sorts of distinctions. Your character doesn’t speak Gaulish, and none of the Gauls speaks Latin. One of your fellow scouts, Tivus, speaks Gaulish at 50%, having come from a northern area of the Empire. The leader of the Gaulish pack has a very high intelligence score, so despite the fact that he doesn’t speak Latin, his intelligence gives him a 10% communication bonus, so he’ll know to phrase simply. This gives Tivus a 60% communication level with the Gaulish leader.

The player of the Gaulish leader sees Tivus’s greeting, and notes an indicator that tells him that Tivus’s pronunciation accuracy is a little over half. He knows that if Tivus does not understand his words, they will get translated wrong, and that less translation failures will occur if he uses short words. His player types:

“This is our home. We want you to leave. We will not let you pass, but we will not pursue you if you leave.”

Tivus sees an interface indication that this is Gaulish, and it reads “This is aguli hom. We want vetat to leve. We will not can you passe, but va will not egak you if yu leave.” This is because the text parser is going to screw up some words, and mildly alter others to represent that 40% lack of communication. Had he typed “Leave our territory at once, you are not welcome here!” it might have come out as “Leav our ekavah at on, you ae not korveg heree”, which is still intelligible, but less so because the longer words have a lesser chance at proper translation.

Tivus uses the interface to indicate he wishes to speak Gaulish instead of Latin. He says to the Gaulish leader “We are few. You are many.” he then points to a bluff in the direction from which your party came. “May we rest there for a while before we leave?”

The Gaulish leader sees Tivus’s text, but with the same degree of inaccuracy that your friend saw the Gaulish leader’s at. The other Gauls see it a little less accurately, at 50%, since they lack their leader’s intelligence bonus. You, however, see both exchanges as a person with 0% skill in Gaulish. Since the leader has a 10% bonus, your overall level of communication with him is 10%. The best you can hope for is a word or two to be correct, and you will understand none of what has transpired until Tivus tells you…

…or, of course, until the Gauls decide that you’re just an advance scout troop for the Seventh Roman Legion and decide that their interests are better served by charging you and bashing your skulls in rather than allowing you to press for time.

Written materials might need to be unreadable except at a certain (high) level of communication between two parties, or translated once only to avoid allowing multiple translations in order to eliminate the errors.

Language training can be another offline character development goal, and this degree of separation helps to eliminate the “casual internet chat” environment of many online role-playing games while providing a framework for upholding character race/nationality differences.

Internet apps like ICQ and IRC minimize the amount of separation you can expect to accomplish. For 3D accelerated games, it’s simply not as convenient to switch back to the desktop to check messages, so the effect in these games is not very noticeable. In any event, preventing users from switching back to the desktop to prevent such programs (or others) is an undesirable goal: it’s your game, but it’s your customer’s computer.

Context, Part I

Context is crucial. Through context, you convince your players that you take your own game seriously. Out-of-context elements in your game are red herrings. Players introduce out-of-context elements all the time. One of the best ways to avoid this to any degree is to lead by example: if your players see you adding out-of-context elements to the game, they’ll not see any reason to avoid doing so themselves.

Common pitfalls include:

• Support characters with silly names: GM Fruitloop, GM Fatality
• Pointless references to real life.
• In-game celebration of Earth holidays: Your world is your unique creation. Litter it with holidays that would make sense in-game. While some midwinter festival has been a common feature in many cultures in our world, there’s no need to assume that a decorated pine tree or gift-giving be an integral part of it. Most nations have a distinctly national holiday, but not necessarily in the middle of summer nor involving pyrotechnics (or arcanotechnics). Most cultures have celebrated a springtime holiday of rebirth and renewal, but they aren’t required to involve rabbits, painting eggs, pastels, and chocolate. We know what to celebrate in our real-life cultures; use holidays in-game to illustrate the uniqueness of your game’s various cultures.
• Anachronisms: Cannons in a world that hasn’t invented other firearms of any sort, let alone sophisticated artillery. Combustion engines powering a flight vehicle in a world that has no history of combustion technology.

Much of that is merely keeping your game in line after basic design decisions have been made.

Perhaps more interesting than keeping context is how to establish it in the first place.

Context, Part II

The current crop of games are all set in a medieval-like fantasy world where magic is commonplace. UO and EQ share many common D&D fantasy/mythological monster clichés: dragons, undead, trolls, orcs, etc. AC is set in a similar world, though most of its monsters are unique rather than being drawn from the same cast of characters as most other game. Even so, some are very similar: drudges resemble orcs, gromnies resemble small wingless dragons, banderlings resemble gnolls, and monougas resemble ogres. EQ is even more steeped in D&D tradition than UO, with a full complement of pulp-fantasy elements such as dwarves, elves, halflings, and the lot.

It would be good for the genre for someone to think differently. And this doesn’t mean abandoning a D&D cliché to adopt a sci-fi cliché. It doesn’t mean combining the two to make the next wave of “steampunk” games, and it doesn’t mean emulating cyberpunk RPGs.

You can make a fantasy role-playing game that avoids clichés. Honest!

Where does your world take place, and what sort of game are you making?

• High fantasy: Here’s a bit of advice: “It’s been done.” High fantasy is usually the “traditional RPG” with humans, demi-humans, dragons, undead, orcs, trolls, and magic that is about as rare as an automobile is in our world. If you’re going to do high fantasy, there’s going to be little to fundamentally separate your offering from UO, EQ, and even AC, SB, and UO2. Strictly speaking, high fantasy simply indicates the author/developer’s attitude toward magic. If you’re going to use high fantasy, you’d best be original about it. A campaign that took place a properly horrific Hell, given a grand scale and opportunity for advancement beyond the lot of a cursed soul, could provide plenty of opportunities for a truly unique MRPG experience.
• Low fantasy: This sort of world has been explored more in fantasy fiction than in RPGs of any sort. The definition of the genre revolves around the treatment of the supernatural. Low fantasy generally has few “fantastic” elements about it: magic is not an everyday part of life, there are less “supernatural” elements (creatures with magical powers as well as spell phenomenon), and that which does exist is often toned down to the point where it is never really demonstrated in a way where it must be presumed to exist: much like the common attitude toward Medieval Europe, most educated people consider magic to be superstition, though many may believe in it fervently. Low fantasy tends to involve stories about people and their interaction rather than epic quests against evil, though much Arthurian literature is a good example of “epic low fantasy”, often involving no magic at all, or at most, magic whose existence is not presented to the reader as an established fact. If high fantasy is the world of the demon-fighting elf paladin with a mystical sword and a bevy of spellcasting followers, low fantasy is the world of the weathered mercenary who needs to decide if his stomach is more important than his morality.
• Historical fiction: Think of historical fiction as low fantasy set on Earth. Historical fiction has not been dealt with in the major MRPGs. Historical fiction would involve more research than creation: if the whole world is simulated, at what time? What sort of conflicts existed, and who were the major players then? If you set the game in 100AD, is your game centered around Rome, or do you simulate the Americas and the Orient and the dominant cultures there, too? Your historical research will come into play: plate mail, muskets, feudalism, central Christian churches, and crossbows did not exist in 100AD. What’s more, common stereotypes and historical revisionism need to be avoided. The Gallic Celt of 100AD Gaul is markedly different than the Pictish Celt of the Scottish eastlands, and portraying them as happy-go-lucky nature-loving primitives would be shoddy research at best. Without your fantasy monsters as a cash spigot, you’re going to have to use other motivations: war between established foes, internal warfare of political treachery, exploration of newly discovered lands with hostile inhabitants, and so forth. Given the Roman attitude of “barbarians”, ranging anywhere from those considered unhuman and approachable only in terms of warfare to those “barbarians” who were assimilated into the Empire itself, even if after a history of conflict, there is clearly plenty of room for differentiation.
• Historical fantasy: Or perhaps “historical fiction” with a fantasy twist. What if your game took place in Europe in the Dark Ages right after the discovery of codified and reproducible magic, alchemy, witchcraft, and/or demonology? How about a feudal Japan under similar circumstances? If a game based on historical fiction seems to be too hard a sell (I personally think, done right, it could explode), this might be an alternative.

Thought needs to be given to creatures. How similar is your world to Earth? Are familiar Earth animals present? If not, what lives in the niches occupied by various mundane creatures?

Developers get caught up in simulating a “virtual ecology” in a world where the constants are much different. How many pigeons would you see in our urban areas if birds were killed for feathers to make arrows? In a world where some degree of warfare or hunting is undertaken by the majority of the population, the answer is “very few”. Herbivores are going to have to be much smaller and resourceful in areas that have fewer plants. Animals can be much bigger in areas that have superabundant prey. Whales eat krill swarms, brontosauruses grazed in water filled with vegetation, “big cats” can only survive in areas with large amounts of prey. This isn’t to say that you should have an ecologically accurate thousand rabbits for every wolf: your players are going to be less interested in rabbits than wolves, no doubt, and more interested in “marsh lurkers” than wolves. Rather than fret about accuracy, use “hinting”: in areas with no (prey), there shouldn’t be any (predator). Don’t use your server’s valuable resources keeping track of how many rabbits wolf A has eaten. In areas with abundant (prey), there should be more (predator), unless (predator’s predator) are in great numbers nearby.

The artist Wayne Barlowe has written books about alien worlds. Thought is given to subjects like biospheres that make sense, issues of prey and predation, and to the individual creature’s place in its environment. There are many lessons in this “whole environment” approach to world design.

If nothing else in a world is near human, how did humans get there? Magical transport from Earth or other unnamed lands seems generic and unsatisfying, and encourages players to bring the same characters they’ve played in other games into your game. In UO, humans came from Sosaria as well as Earth. When players bring in their existing character personas, they dilute your world. The solution isn’t to lay down the law on your players, the solution is to provide them with alternatives that make them want to play in your world, rather than just show up.

Context, Part III

Everything your players can do in your world must be make sense to the fiction. Repeat this phrase often. This is one of the ways you give your world context. The existing games have serious flaws in this category:

In Ultima Online, one of the biggest offenders in this regard is resurrection. Especially disorienting is the way the game fiction treats death. When a fiction-centered NPC dies, that’s it - he or she is dead. There might be a funeral, a search for the body, or mourning. When a PC dies, his or her ghosts jogs over to the nearest wandering healer or shrine and shrugs it off. While the idea that “death should mean something” has validity, the simple fact is that if death is anything other than permanent, the difference needs to be backed up by the fiction.

When Seer or GM-controlled plot characters died, that was it, they were dead. Sometimes, when dealing with players with less experience with plot events, people would effectively scratch their heads wondering why the character didn’t resurrect. Other ghosts would follow that ghost and try to help them to the nearest healing point. When the post-event fiction was written up, death was death, and never explained otherwise. In events and plots that I had a hand in, I always strove to provide some reason to explain this. Sometimes, it was sufficient to explain it with the concept that dark “soul-banishing magic” was used which prevents the soul from re-entering the body. Often, this was enough, and the players were more likely to respect the outcome without going back to their guilds bemoaning the silly way in which events treat death.

In Asheron’s Call, any player has the ability to send instant communication to any other online player in the game. By using the /tell command and other features, you can instantly message people by name, your patron, your monarch, or your fellowship. With the friends list, you can tell the moment that character is logged on. Rather than cave entrances to underground dungeon environments, often the entry is a shimmering portal on a plain. While the portals are explained in the fiction (and are, in fact, central to it), the instant communication is not. The only way one can logically explain it in the context of the game is to attribute it to some sort of psionic power or other telepathic ability, but no such explanation is offered. To have that communication without an explanation of how it takes place gives private conversation in AC a very Quake-like feel.

Explaining the conveniences of your game system in your fiction is your job, not your players’.

The Law: Players

Ideally, your game world would be treated as such, and your players would have a common sense approach to decency, fair play, and what does and doesn’t constitute appropriate behavior…

…and if you design with that assumption in mind, you’ll be in for a rude awakening.

You need to make very clear what is and isn’t acceptable behavior, and enforce it diligently. This doesn’t mean you have to have a draconian set of policies; this simply means that whatever your policies are, you need to communicate them very clearly to your players and the staff which enforces them.

Contrary to the UO model, this needs to be kept in mind during the design phase rather than ignored by programmers as a support responsibility that takes effect after the development work is done. Adding accountability on top of a structure not designed with such concerns is a difficult process, needlessly so. Give your support team the tools they need to properly administer the game as soon as it’s out, don’t put it on the back burner.

AC has taken a line on exploitation that neither UO nor EQ has tried: programmer accountability. When Turbine releases buggy code (as every programming team does from time to time, this is the nature of the beast), this is treated as a problem with Turbine, not a problem with players, and steps are taken to resolve the issue without punishing players for problems they didn’t cause. OSI has usually seen fit to ignore its mistakes until something brings it to a head or until the player community can scream no louder, and Verant is more apt to hold its players responsible for avoiding any activity which even resembles AI exploitation, let alone willful exploitation by players. This often puts OSI and Verant (even more so) at war with its players, which poisons the relationship of the development staff and the customer.

While a “zero tolerance” program might make sense in situations where, due to game design, a certain exploit is unpreventable (there will always be a dupe bug in UO unless OSI makes fundamental changes to the way it tracks characters across server boundaries, which simply isn’t practical), Turbine’s model seems more desirable for situations where a code-based solution can be coded into the next patch. Turbine’s publisher, Microsoft, allots Turbine a generous and predictable patch size allotment per month, which has given them flexibility in this regard. If your patch allotment (or your development staff size) is small and you’re not ready to make that sort of commitment to closing exploitation loopholes, it is that much more critical that your support and development staff have a very clear picture of what is and isn’t exploitation, harassment, and violation of your usage agreement. UO in particular has had a poor track record in this department, and has had problems since the beginning with GMs disagreeing on what is and isn’t a violation of their TOS.

The Law: Characters

Danger needs to exist, and many of your players are willing to dole it out. The problem is that if you make it too easy, you get remorseless killing without consequences.

In a large world, there is room for different societies with different laws. Perhaps much of the “civilized” world has similar laws against murder, thievery, and crimes against the state (leading insurrections, for example). A character that lives in one kingdom and who kills a soldier of the king’s guard will almost certainly be wanted dead by the authorities in that country. In an enemy nation, however, they might consider such an act a heroic deed and welcome him - as long as he confines his killing to the subjects of his former king rather than the citizens of his new home.

Some areas may be completely uncivilized and not have any sort of law enforcement. Others may be so different from the norm that they become a haven for criminals.

However this works out, there needs to be a means to keep criminals out of areas in which they are no longer welcome, and the consequences for risking being caught in those areas must be severe. A militia of NPCs and PCs may have an order from the King that allows them to hunt down known criminals. If this is at all practical, it’s much better than invincible guards popping out of nowhere. There’s a completely different feel to a player when a character gets whacked by an instant-kill guard the moment he crosses an unseen city boundary and that of being hunted down by an angry mob that grows the longer it’s evaded. Maybe the odds are still miserable, but if there’s a slim chance that, properly executed, freedom can be had by escaping the kingdom or city, that’s more entertaining. In feudal and pre-feudal societies, the law wasn’t apt to “forget” the wrongdoings of criminals, either, so in cases of serious crimes like murder, the guilty character shouldn’t be able to stroll back into the capital after a few weeks of beating down “evil” creatures or persons. Only in rare cases should this be possible. For lesser crimes, the law isn’t as apt to remember the offender as long.

Regardless of how death is treated in your game, the penalty for an “outlaw” character getting caught by (or brought to) the law should be significantly greater than the penalty for normal death at the hands of a personal enemy or a monster.

If players are able to form communities with their own laws, members of the law enforcement arm of that community should be able to flag a character as a known criminal. This opens up possibilities of misuse, but given that player communities are generally small and easily avoidable (cities don’t move), each will in time earn a reputation among other players, and lawless cities aren’t going to attract as many merchants, though they may attract more mercenaries and those looking to hire them.

In any case, there should be places where the most vile criminals are not hounded by the law. Areas like this are dangerous, and anyone entering them should be prepared for danger.

Conclusion

The MRPG development community is going to find out very shortly if there is enough room in this market for five or more concurrent offerings. I believe that the first title that represents a significant evolution over the current works available (and several in-progress) may have a chance to grab a majority market share from a currently-fractured player-base spread across several games.

UO was ambitious even for its flaws. EQ attracted an audience that preferred a 3D environment or who had not found UO to their liking. AC attracted an audience looking for the next “good thing”. The titles in development are going after a finite number of users who will play more than one game concurrently, or are willing to switch from one to the other. I think they’ll find that there aren’t any more “free rides” in this genre…

…it’s time to innovate again.

Delusion