This is the fourth 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, or unless I should happen to be distracted by a pretty butterfly or something. Which is exactly what happened after I wrote Part III of this series in July 2009!
In this long awaited (right?) Part IV of the series, I’ll concentrate on the following law:
- Macroing, botting, and automation
Macroing, botting, and automation
No matter what you do, someone is going to automate the process of playing your world.
Corollary:
Looking at what parts of your game players tend to automate is a good way to determine which parts of the game are tedious and/or not fun.
A background lesson about automated gaming
Bots. Botting. Macroing. These terms incite a lot of emotions in both players and developers of MMORPGs. Ever since the days when MUDs were the epitome of online RPGs, there has existed software that allows someone to automate gameplay either to avoid having to do the boring parts of the games by hand, as a quick way of reaching the “end-game” or to amass piles of gold and wealth that would take a normal player days, weeks, maybe months to achieve.
This is possible because (most) human players normally have to take a break every now and then to rest, to socialize/deal with friends, family and loved ones, to sleep, to eat and so forth. A bot – essentially a computer controlled player – on the other hand, can go on and on without pause, without the need for food or drink, without having to rest. This of course makes them exceptionally well suited for tasks such as, say – killing one million boars in an MMORPG for loot, experience points and rapid character progression.
These days, most people think of gold-farmers when you mention “botting”, but my first experiences with the topic at hand took place in Ultima Online in early 1998 and wasn’t with gold farmers, or completely automated bots – but with macros. Some players would record certain activities in programs such as EZ Macros and then use those programs to replay the activities over and over again without having to sit there and actually push the buttons themselves, or even monitor the process. These activities included stuff like stealing from a pack-horse filled with worthless items over and over again, sailing a ship in circles while fishing, or mining ore from the mountainsides and depositing the ore in a packhorse or in a nearby house when the character couldn’t hold any more ore.
This was beneficial due to Ultima Online’s usage-based skillpoint-system, where you had a chance of gaining a skillpoint in a given skill every time you used that skill. So anyone who automated that process would naturally get the maximum amount of possible skillpoints in a skill in a much shorter time than those who did it by hand.
The “AFK macroing” in UO seems insignificant compared to the more advanced botting programs one can see in modern MMORPGs today, and in a way that is true. The macroing that took place in UO was more an indication of a bad/broken design than anything else, and with the exception of resource-harvesting and grinding up skills to the cap (which was relatively quick, even for non-macroers) it was somewhat limited in scope. More advanced “AI” has been added to botting programs since then, though. Bots in todays games can more or less play the games by themselves; they take predetermined paths through the game-world, react to scripted events and triggers, drop off loot at storage vaults when inventories are full, join PvP battles for the rewards those can give, and so forth – with no need for interaction from the actual human player apart from maybe being around to respond to “Are you a bot?”-like inquiries from Game-Masters.
So what? Does it matter? Why should we care?
“Does it really matter if someone automates playing parts of the game? If they just do it to bypass the boring parts of gameplay, and it doesn’t affect me directly, why should I care?”
Depending on the type of activity being automated, it can have some very negative consequences . Also, even if it does not affect any one player directly, it will affect all of them indirectly. Here are some scenarios outlined to support this theory:
- Botting promotes gold farming and destroys the virtual economy
Gold farmers are botters who automate playing games for one specific reason – to accumulate resources (wealth, resources, items) as quickly and as efficiently as possible, and to then sell these virtual goods (usually in the form of in-game currency) to other players for real money. In the process of doing so, they not only spam the in-game chats with advertisements for “cheap gold!” until regular players are swearing and cursing at them left and right, they also flood the virtual economy in the game with resources, to the point where those goods are no longer worth anything for normal players trying to make a virtual living. This is not just true for gold-farmers, but also for normal players who automate processes that let them harvest resources, produce items or bring wealth into the world faster than a normal player would otherwise be able to. Gold farming also leads to another problem, which is that players’ in-game wealth might depend on their real-life access to excess funds which they use to buy their way to the top, as it were. - Notion of a level playing-field destroyed
Being able to buy one’s way to the top with real life money aside, botting/macroing that automates the processes through which a player-character advances in a game will in general give unfair advantages to those players who do this, and will destroy any notion of there being a level playing-field for all players. If a small subset of players have these advantages over other players in a game with a very competitive setting, this might in turn lead to players losing their motivation to play the game, as they might feel (and rightly so) that continuing to play is an exercise in futility – they can never keep up with or compete with the botters. - Unbalanced Player vs Player battles
Consider an online game that pits two teams of 50 players against one another in a large battlefield. One team consists of 50 active players – the other team consists of 30 active players and 20 players who are running botting software in an attempt to farm whatever currency or reward that this player vs player activity will produce. At best these bots are just about smart enough to follow very specific predetermined patterns of movement/actions, but are unable to react to the unpredictable nature of the players on the other team – or even to support their own team in the event of something happening that is not covered by their scripting. At worst they join the game and are stuck running against an obstacle for the duration of the match, contributing nothing to the team and taking up a slot that could otherwise have gone to a legitimate player. - Botting is disruptive to player/developer relations
In any game where bots are allowed to run rampant, unnecessary tension is generated between players and the developers of said game – especially if the developers appear to be ignoring the problem through inaction. The longer this inaction (perceived or not) is maintained, the stronger the tension between the players and the developers becomes, until it reaches a point where the players feel that the developers no longer care.
Is bad design to blame for botting, or is this a problem that cannot be fixed?
“The law at the top of this article states that players automating the process of playing the game is inevitable. Does that mean that no matter what we do the outcome is the same, that there is no point in trying to address these problems?”
Botting and macro usage is inevitable, yes. Would players still do it if the game in question had been designed so all aspects of the game was equally entertaining and/or interesting? Probably not to the same extent.
However, while there are certainly players who will only use these kinds of methods to avoid being afflicted with/already suffering from carpal tunnel syndrome, or to just get past all the boring bits, there will always be players who will seek to gain any advantage they can over the competition. If they can automate processes in the game (whether fun or not!) by using botting/macroing software to become more efficient, advance faster and in general out-pace the other players, they will.
Online games/worlds have tried a variety of methods to combat the problem, either through mechanics that try to directly detect bots, indirectly hamper/create problems for non-human players or through systems that attempt to level the playing-field – for instance by introducing caps on just how efficient a player can become, or by giving players tools they can use to keep up with other players.
Examples of the former includes automatically detecting and flagging players as bots based on analyzed data, random mob spawns during resource-harvesting (probably more frustrating for normal players than for bots) and mechanisms for reporting suspected bots (relies on players actually reporting and on the administrators of the world actively responding to those reports).
Examples of the latter includes offline systems for training skills, feats and/or levels, rested bonuses that increase your character’s efficiency when you return to the game after having been offline for a while and dynamically adjusting the frequency at which loot drops from mobs based on how fast (or how regularly) they are being killed.
As these systems are put into place and evolve, though, so do players and developers of botting softwares!
Conclusion
Whether to combat non-fun bits of games, to gain the upper-hand against the competition or to make a living through selling gold to other players – there will always be people who will attempt to automate parts of the games they play through botting software and macro programs.
Developers can put roadblocks in their paths, can go on banning sprees, can publicly proclaim that the game will shut down unless botting comes to an end – yet botting will still happen. Ultimately this is because people are broken an arms-race between the developers of the game and the developers of the botting software, and only sustained efforts by the developers can keep the amount of botting going on in their games to a minimum.
Finally, remember the following, kids:
- Every time you bot or macro your way to perceived victory, Sid Meier kills another game design with dinosaurs in it! :(
Great post – got me thinking about the good old days of macro-ing/botting. As a “criminal” of the Online game/MMO world (no, I don’t do it anymore, I am too old to be bothered) I don’t think botting is that bad tbh. From my point of view, it boils down to smart people exploiting chinks in the design to gain an advantage for their personal pleasure. By revealing design issues it helps companies fix their problems, and players to expect a better game. The problem arises when such scripts are released to the masses, or when people start multi-boxing their scripts – that’s when entire game economies can start to break down. That distinction aside, There are three major approaches companies can take to mitigate the negative effects of botting:
1. Fight it by detecting botting software, keypress sequences, server communications, injecting randomness, admins, player reports, etc. This is what most companies have been doing so far. It is the most straightforward, the costliest, and probably the least efficient method to prop up a weakly designed economy.
2. Change your design so that there are no bottable tasks – This is what I believe Raph Koster is advocating. E.g. if the only way to get at resources is to fight actual battles with NPCs or other players, there is no way bots can be more efficient than a highly trained farmer (IMO a highly trained farmer = a good customer & a proficient gamer who pushes other players to improve their tactics). Even if no fighting is involved, how about having resources that move around? (e.g. rabbits that you have to chase before you can get their fur, etc.). As long as it makes sense in the context of a game, it is not disruptive. E.g.
Example A) Bandits jumping out of a rock when you mine it : BAD
Example B) In a space-trader game, having to fight cargo/pirate ships to get resources : GOOD
3. Fight botting, by creating a semi-realistic economy. See, in an MMO e.g. you can click a rock to get an iron ore. Then you can smelt the iron to get an iron ingot which you can sell for a profit and if you find a way to automate this process, you are a made man. This is NOT how the real world works.
a) In the real world it takes skill and knowledge to discover ore in the first place. It also takes knowledge to build an efficient mine/extraction method without killing yourself in the process. This is beautifully replicated in Minecraft when people have written mathematical analyses on optimal mining (http://www.voxelwiki.com/minecraft/Elites_of_Minecraft%3a_The_Miner)
b) In the real world, automation is more complex than repetitive low-yield actions. If you want a unit of sand in the real world, you need a shovel – if you want 1000 tons of sand, you either need 10000 people working for you (with pay, organizations, etc.) or you need a few bulldozers and loaders (which are hard to operate, and need tons of engineering, blueprints, materials, etc.) but are a LOT more efficient. Traditionally, games will not allow you to create/operate bulldozers – hence the 10000 bots people come up with. Assuming that it is possible to have the equivalent of bulldozers, wouldn’t this ruin the economy by flooding it with sand? NOT if the sand can be put to good use. NOT if there is a huge market for sand. Traditionally, games use the sand to make other items, and so on which culminates in making the Uberweappon. This is ridiculously short-sighted. In the real world, no amount of sand or other basic materials will allow me to create a stealth bomber – I need knowledge – knowledge that in itself is compartmentalized, licenced, and secretive. Imagine e.g. that creating an uberweapon involved 10 different items, and each of those required 10 different materials. Also imagine that getting those materials needed real skill and making each item needed a license or scroll that needed to be licensed (rented from the system as it were for a lot of money), and that different guilds or people had ownership of a limited number of those licenses because they are expensive to rent, limited to N per server and because they are lootable from the corpse of the license holder – new game mechanics would evolve around complex trading/sub-licensing/politics/assassinations, etc. agreements that bots can’t really handle.
c) In the real world even the best miner CANNOT own/fly/operate stealth bombers and nukes. It is not necessary for good gameplay to give all players the SAME opportunities, Rather you should allow each player to evolve in the direction of their choosing. You should be able to be an ace pilot without having to own an F-35! In a game this would translate e.g. to: Separate weapon building skills from weapon usage skills so that players have to pick between being soldiers (who use weapons), blacksmiths (who build weapons), wizards (who use spells) and alchemists (who prepare spells), etc. The finer the breakup, the more interesting the economy and inter-player relations.
There are many other ways in which a real-world analog can help us design robust MMO economies, but there is also one painful lesson to be learned from the real-world:
DISRUPTIVE TECHNOLOGIES ARE GOOD!
The introduction of the automobile forced a lot of inflexible horse breeders out of business – yourgame economy needs to allow for disruptive technologies by merit of being complex, requiring huge amounts of cooperation and constant monitoring by designers who know about economics. In the real world, e.g. just owning a car that can do 200 mph doesn’t give you the right to speed on urban roads. If you do, you will be punished by the families and friends of people you are bound to run over (or by the State as it were). This highlights the need for a minimal amount of realistic physics, and a political structure in which a community of players can organize themselves and by doing so control the fate of the world and other players, but that opens up a whole new can of worms some designers would say – I would say it is a challenge that others tried to meet, and some did it better than others. Whatever happened to developers taking on big challenges and developing solutions anyways?
The issue of designing a persistent game world with an organic economy got me thinking about the good old days of AI research. The problem of creating a complex open-ended organically evolving world is very similar to the problem of creating a learning “human” AI in many ways:
1. Both involve a deep understanding of the human brain
2. Both require knowledge human motivation both in terms of personal and societal needs.
3. Both are complex systems made up of simple operators following simple rules.
4. Both cases involve huge amounts of parallel information processing and information transfer between the simple elements.
5. Both are associated with a lot of false dogma and a common conception that it simply cannot be done because humans are “special” and that “simplifying humans of the world” and writing them out a set of reductionist algorithms necessarily breaks something.
6. Tackling both problems requires a fusion of knowledge among different disciplines such as physics, computer sciences, game theory and psychology.
7. So far, we have failed to achieve both. On the AI front, there are breakthroughs every day – on the persistent world side most designers seem to have accepted defeat and have been backpedalling for a many years, especially when MMOs are concerned :(
Great replies, Babak – either one would be good material for stand-alone posts themselves (though it’s possible you already have)! :)
I agree with most of what you say (how boring of me, I know…), but concerning 3c) in the first comment. Specialization like that can lead to a different set of problems again – “mules”. The big challenge in that context is how to encourage those inter-player relations vs players just creating enough characters to cover all their needs themselves. I guess if the skills were complex/deep/challenging enough that they would be hard pressed to excel at all of them (even using separate characters), it could work.
And as for the following question… “Whatever happened to developers taking on big challenges and developing solutions anyways?”… what happened was that the bosses and/or publishers of those developers stopped taking risks because….because… I don’t know…. “WoW doesn’t have that/do that, so we don’t need it either”. Most innovation these days seem to come from independent developers who _have_ to take risks because succeeding at that is one of the major ways they can get traction in the press/amongst gamers.