05 Mar, 2012, Deimos wrote in the 41st comment:
Votes: 0
@Rarva: I don't really understand what point you were trying to make, but social behavior is not an advantage in anything we've discussed so far (if you're using the word "social" in the proper manner, anyway).

@KaVir: Ultimately, I would like to render botting useless (or as close to it as possible). That's why I phrased it that way. I don't see that any of your examples significantly reduce the utility of botting. At least, not unless you're comparing them to the worst-case scenario, such as IRE combat.

To give an example, your primal scenario rewards botters more than average players. While you're looking at the situation from a time spent-to-rewards gained ratio point of view, a botter looks at it from a manual effort-to-rewards gained ratio point of view. And by that measure, a player can kill 50 mobs for 15000 primal, while a botter can go to Starbuck's and be gaining 60% more primal for zero manual effort.

Yes, you've reduced the gap between bots and players, but you've effectively done nothing to reduce the usefulness of botting. Players by and large don't bot to attain what normal players *can't*; they bot to attain what normal players *can*, with little or no effort. At least, that's always been the case in games I've played.
05 Mar, 2012, KaVir wrote in the 42nd comment:
Votes: 0
Deimos said:
@KaVir: Ultimately, I would like to render botting useless (or as close to it as possible). That's why I phrased it that way. I don't see that any of your examples significantly reduce the utility of botting.

The first example I cited was a mud in which "Combat is fully automated, so there would be nothing for the bot to do during the fight. There's no character advancement, so the bot couldn't improve you if left running overnight. And fights are always one vs one, so there'd be no benefit in running multiple bots".

You seriously don't see how that would reduce the utility of botting?

Tell me then, what advantages do you think the bot would have over a human player in that example?


My second example handled almost all the tactical and speed-critical decisions through preconfigured options, so you didn't need to do anything during a fight (as fights were generally extremely short anyway). Character advancement was based primarily on playing time, and if you wanted to idle for an hour each day you hardly needed a bot for that. The only time bots would have a clear advantage is if you created a gang of them, but that's more of a multiplaying problem than a botting problem (you'd have exactly the same issue with a player who was controlling multiple characters).


Even my third example has greatly reduced the value of botting, although the solution is rather more complex due to the nature of the game. I can understand how you'd have trouble wrapping your head around it, but it does demonstrate the fact that the more complex gameplay, the more difficult it becomes to reduce the value of botting.
05 Mar, 2012, Deimos wrote in the 43rd comment:
Votes: 0
@KaVir: Something must be getting lost in translation. Even "automated" combat a la Diku-style allows character input during fights that affects the outcome (kick, bash, spells, whatever…). If you're *really* saying that combat in that game allows no input that would alter the outcome of the fight, then I certainly won't suggest that botting could somehow be useful in that situation, but I highly doubt this is the case, as that would be a horribly boring system. That leaves the alternative, which is that combat is only partially automated. And, yes, even in that case botting can be a significant advantage for above cited reasons. Sure, you can design a system where botting is rather pointless, but it would seem to also require that combat be pretty simplistic, boring, and ultimately not worthwhile.

You also seem to have skipped over my point about the difference between reducing the gap in capability between bots and humans, vs. actually reducing the usefulness of botting. I thinl it's a very important distinction that you seem to be overlooking. Just because a bot will never be "more skilled" than a player does not mean botting is less useful. I even demonstrated why this is the case using one of your own examples.
05 Mar, 2012, KaVir wrote in the 44th comment:
Votes: 0
Deimos said:
Even "automated" combat a la Diku-style allows character input during fights that affects the outcome (kick, bash, spells, whatever…). If you're *really* saying that combat in that game allows no input that would alter the outcome of the fight, then I certainly won't suggest that botting could somehow be useful in that situation, but I highly doubt this is the case, as that would be a horribly boring system.

Yes, I'm saying that Gladiator Pits has no input during combat. I didn't have enough space to add any combat commands, so I decided to turn it into a real-time version of one of those play-by-email fighting games, where you design your warrior in advance, then receive descriptive reports of their fights. In this case you actually see the fight as it unfolds, but the concept is the same.

You asked for examples of actual implementations, so that's what I provided.

Deimos said:
That leaves the alternative, which is that combat is only partially automated. And, yes, even in that case botting can be a significant advantage for above cited reasons. Sure, you can design a system where botting is rather pointless, but it would seem to also require that combat be pretty simplistic, boring, and ultimately not worthwhile.

No, the combat in the second and third examples is more interesting, and each provides different mechanisms to greatly reduce the value of botting in a combat sense - by that, I mean in a fight between a human and a bot, each with characters of similar strength, the bot does not have any significant advantage.

This should be viewed separately from advancement, because that is a separate game mechanic entirely. The second example I provides is a good example of an advancement system which can't reasonably be botted - I guess you could set up a bot to kill 100 mobs, but that wouldn't take more than a couple of hours and then there'd be nothing left to bot.

The third example is significantly more complicated. In theory the bot can advance faster, but the more it kills the easier it is for other players to find, and if they kill it they'll steal increasingly large quantities of its primal. Furthermore, as more and more quests are added, the reliance on regular killing decreases - my long-term goal is to have a game in which you can progress almost entirely through quests (which only give their reward the first time you complete them).

At that point it would be theoretically possible to remove other forms of advancement entirely, which would make it rather pointless to advance through botting (because you'd have to design a different script for each quest, and you only get the reward once per quest, so it would be faster just to do them yourself).

There's also a fourth example: Gladiator Pits III. This used exactly the same combat system as the third example, but had no form of character advancement - it was pure PK. Combat was also based on challenges, so if a player decided to create an army of bots then people could just refuse to fight them.
05 Mar, 2012, Rarva.Riendf wrote in the 45th comment:
Votes: 0
Quote
@Rarva: I don't really understand what point you were trying to make, but social behavior is not an advantage in anything we've discussed so far (if you're using the word "social" in the proper manner, anyway).

What I mean is that the human can have info a bot cannot have (or it would be way too impractical to code them all) like social ineraction between players, and those are way more valuable in fight than just to know the exact second a spell will wear down (and that in itself is mostly irrelevant in most of the fight you can engage in the game I develop, as the fights between humans are usually either very short, so you have no state known anyway (or long enough you can choose to flee pretty much at any time)).
A bot is clueless and will perform poorly in pk in my game cause he misses those crucial info like 'is this a trap from a player of a clan that is at war against me' instead of just 'this is a pure one on one' fight. Again it depends on your game deisgn. Though if you jsut want a bot that never dies and jsut flee…feel free. Pretty much useless though.
05 Mar, 2012, Hades_Kane wrote in the 46th comment:
Votes: 0
KaVir said:
Hades_Kane said:
We disallow botting as well, but we pretty much define botting as having these things automate your character without the player present. If someone has automated their character and are watching the screen or, for that matter, has a split screen and can respond to someone in a relatively quick amount of time, then that is ok.

"Don't turn your back, don't look away, and don't blink."

Sounds like response rule rather than a botting rule though. It'd be simple enough to set a trigger to play a sound when someone sends you a message. On the other hand, someone who wasn't using triggers might miss a message in the heat of battle, with all that text flowing up their screen - and they'd presumably be "punished" for not responding.


Generally speaking, we look for signs of a trigger first with this. If we can reliably trigger someone's setup, they aren't responding, they normally get a ridiculously obvious, big block of text as a last warning kind of thing. You can't really miss it reasonably. If someone still fails to respond for a bit of time, they'll normally get transfered to "The Island" (while on this island, no exp can be gained, no skill gains can occur, basically anything that can cause a character to receive a positive benefit is turned off here), and then when they return to their character and notice themselves out of place (often times after an hour or more) then we'll address the situation with whatever measures is considered appropriate.

We took our smite command, for example, and rather than that just killing someone and junking their stuff, I implemented "smite levels" where the first level is basically the warning label, this is getting caught doing this one time and they are let off. Each subsequent use of the smite command increases their level and carries with it automated measures. This has become a popular method for dealing with idle botters as it allows the immortal to have to make very little in the way of judgment calls for as for what is appropriate, and players understand clearly what the steps along this path are.


As far as "scripted assisted players" I have no problem with that at all. As long as someone is actively at the game and has some measure of control over their character and can respond when necessary, they can script however they want. Triggers for disarm recovery, for example, is one popular form of attempted scripting. One thing we have done with our game, though, which makes it harder to bot (and is overall a smarter game design decision overall, I feel) is we don't have "lag" for skill use. Where as on a normal diku MUD scripting {get weapon;wear weapon} would be useful when you see 'has DISARMED YOU!' because it adds to the command stack, on our game when you use a skill or otherwise have any action that would cause "lag" elsewhere, you are instead entered into "wait time", and while in wait time, any other wait-restricted commands (such as bash, kick, get, wear, etc) simply gives a "You are not ready" message and no command is added to any sort of stack. Non wait restricted commands (such as score, tell, who, gossip, etc) will fire immediately without any sort of wait. This makes it harder to script doing something like picking up your weapon on a trigger. Bot prevention wasn't the point of this, just a happy side affect of helping make that more difficult.

We also try to put a bit of work into scripting many of our mobs for more unpredictable or player-like combat. Mobs that might change elemental weakness in battle, others that might charge up for a big blast, others that might change forms or other aspects that would require a thinking player to adapt… Of course, a tenacious enough scripter could probably tailor individual scripts toward individual mobs, encounters, or areas, but more interactive mob fights also ought to have less of a chance to be easily bot-able.

Overall, though, bot enforcement isn't a priority on the game. The vast majority of players simply respect the "Don't bot" mentality and so it doesn't come up very often, and generally we don't try to "legislate" in place of code. We have a total of 8 rules, for that matter, and the first 6 are basic common sense and conduct rules (don't aid your alts, report bugs, don't spoil other's RP, don't reveal in-game quest solutions, don't advertise other MUDs, watch your language) with the last 2 being the closest to making rules in place of code (no multiplaying, no botting). Even the quest thing in some respects has been addressed game side wise as we have systems in place to set values on players and remember them, and some of our quests have near randomly generated solutions that even if shared with others, won't usually be of any use.
06 Mar, 2012, Deimos wrote in the 47th comment:
Votes: 0
@Hades: And players don't just make a trigger for "You aren't ready yet" to resend the command (spamming your server in the process)?
06 Mar, 2012, David Haley wrote in the 48th comment:
Votes: 0
Rarva said:
What I mean is that the human can have info a bot cannot have (or it would be way too impractical to code them all) like social ineraction between players, and those are way more valuable in fight than just to know the exact second a spell will wear down (and that in itself is mostly irrelevant in most of the fight you can engage in the game I develop, as the fights between humans are usually either very short, so you have no state known anyway (or long enough you can choose to flee pretty much at any time)).

Emphasis mine.

I think it's great that your game has this or that property that has this or that consequence, but this is a discussion about games in general, not only your game. It's not very useful to reply to a general comment that says "botting can be really useful when XYZ is true" by saying "well in my game XYZ isn't true so botting isn't useful". You're not having the same conversation and it just makes things really confusing and it feels like disagreements for no reason and nobody is learning anything. :sad:
06 Mar, 2012, Deimos wrote in the 49th comment:
Votes: 0
@KaVir: WRT your proposed questing idea, it would really depend entirely on what a quest is comprised of on your game. I know plenty of games where quests are touted as "unique", while actually being nothing more than permutations of extremely bottable components/phases. For example, "retrieve X item and deliver it to npc Y". You can vary X and Y, but a bot doesn't care. Programming in the locations (if they're static) or methods for locating the items and NPCs can be trivial in a lot of cases.

And coming up with truly unique and impractically bottable quests is a really tough job IMO. Not to mention an ongoing endeavor. I think it would almost be impractical, and near impossible if quests are the sole method of advancement like you alluded to.
06 Mar, 2012, Runter wrote in the 50th comment:
Votes: 0
Bot aren't better at all things than real people. The case for making a game where botting isn't terrible useful is focusing your game on those things. One way is making the world a dangerous place with serious consequences for flaws in judgment. Most games don't do this, and thus bots have a highly favorable risk-reward scenario. Instead, games try to balance this with the risk of being caught by staff. I think a better way is making the risk inherent in the game in such a way that a non human may not be as effective in navigating the risk. Make the risk/reward the right balance and people may not find it beneficial to do so. And even if they do, maybe they had fun doing it. And maybe they didn't end up with such unmitigated benefits. So it's a win-win and you don't need to police it.

Or don't do that, and make a game where you just type kick over and over and tell people they can't automate it without risk of staff intervention. Which ever one seems more reasonable. :)
06 Mar, 2012, quixadhal wrote in the 51st comment:
Votes: 0
Instead of asking yourself "How do I keep people from botting my game", maybe you should ask "Why do people feel the need to bot my game".

*** Book ahead, get some coffee if you venture forth ***

Why do people bot? Typically, it's because your game is boring and repetitive, and instead of enjoying the process of playing from start to finish, they want to cut out all the boring repetitive crap in the middle and get to the endgame.

If I play a game, and I find myself developing a regular pattern when doing something, regardless of what that thing is… maybe killing boars, maybe harvesting iron nodes, whatever… and I feel that I have to be doing it for X amount of time to continue advancing my character, it's pretty normal to start thinking of ways to do it more efficiently. If X is very large, it's also pretty normal to try and find ways to make the computer do it for me, so I can watch TV or read a book or whatever and save myself the carpel tunnel of having to do it so much.

If that's the case, then you might want to rethink your game mechanics. Does the butcher really NEED 50 boar heads from EVERY player that trots through the area? Just how many bandits can reasonably infest the road to town anyways? If the player just finished a quest to pick 50 dandolions, do you really think they're going to enjoy picking 50 tulips too? After killing the 1237th chicken, do they really still learn as much as they did killing the first 2 or 3?

So, let's ignore text MUDs for a moment, and look at successful games. You know, ones with more than 100 people actually paying money to play them. What makes them successful? Of course, pretty graphics help. But surely that isn't enough.

I'll pick a couple of elements out of some very successful games. Everquest was a huge success, partly because it was one of the first graphical MUDs, but also because it was HARD. When you walked out of the starting area, you expected to die. What's more, you expected to die and lose all your stuff. Walking into a dungeon was creepy, because it was dark and even with a torch, you could only see a few feet in front of you… by the time you saw something, it was already trying to eat you! Getting lost in Nektolus forest at night was almost certain death.

Adrenaline is a must have for a game to be fun. If I don't get scared, or excited, or even angry… why am I playing?

Diablo 2 was also a giant success, because of random loot drops. You didn't flip through a guide and go "Oh, to get the Shoulderpads of Brutal Shouldering, I need to kill Lord Whassits 300 times". Now, you could look in a guide and see a list of mobs which might drop those shoulderpads, but they weren't tied to a single guy. Basically, they had random loot tables for each type and difficulty of critter, and when something dropped, it also would have random attributes attached to it. In short, Diablo plays like a slot machine. You run around killing stuff, and when something rare dropped, it was like seeing the numbers line up and you had to go identify it to see how big the pot was.

Gambling is popular. Feeling like you're winning is fun. Giving people the knowledge that there's ALWAYS a chance to get something cool from doing even a boring or repetitive task may entice them to keep doing it.

World of Warcraft. 10 million people still play it. It's boring, with thousands of "fetch me a spoon" or "kill 20 garglesnakes" or "go talk to foosle in bugville" quests. You level by slaughtering thousands upons thousands of the same things. You grind dungeons over and over to try and get rare drops which will be obsolete in the next patch. Why do people play it? Their friends do.

People like to play with other people, especially people they know in real life. When you're at work, and your coworkers are talking about a game YOU play, you now have something in common to relate to. Even a pathetic grind game can work, if you can get enough people to play it together, because they enjoy playing it together.

EVE Online is a nitch game. By MMO standards, it's pretty tiny… but 250K players still dwarfs the combined total of every text MUD running on the planet by a huge margin. It's insanely difficult and unforgiving. It is a ganker's paradise, where the strong prey upon the weak and laugh. It has very little in the way of variety of content, and as a solo game, is pathetic. Yet, it has 50K people online at any given time, all working towards their own private goals. Why? Because it offers every player the chance to change the world. As a sandbox game, much of the map is controlled by the players, who fight amongst each other and work together in giant corporations to say "This bit of space is OURS!"

People enjoy owning things, and feeling like they matter. As a part of a larger group, you feel like what you're doing is making a difference, and that if you weren't there doing it, your friends and allies would have a harder time of it.

So, let's go back to the topic of botting. How can you make people NOT want to bot your game? First, you need to recognize and eliminate the boring parts that would make someone want to skip over them. That means getting rid of the easy-to-make content that has become staple for us over the years. Lay off the "fetch me 10 bear asses" quests. If the game isn't as repetative, people won't want to bot, and bots won't be as big an advantage anyways. Next, make the game more exciting. Everyone likes seeing the "DING" of gaining a level, but that isn't enough of a reward. Everyone likes killing some big boss monster and getting loot, but if the fight wasn't hard enough, or if there was never any real risk of losing anything, *yawn* wake me up when it's over so I can either grab the loot or reset the instance and try again. Death needs to be something people are afraid of, otherwise there's no point.

Now, balance the risk vs. reward idea with some variation. The players should expect killing something easy should give them junk loot, and killing something hard should give them better loot… but it shouldn't be set in stone. They should have a chance to get something cool from doing mundane things, and also not know for sure what cool thing they'll get. Keeping them guessing helps keep them interested.

At this point, you need players to team up and play together. Don't cater to solo people… they can play solo games by themselves. They came to a multiplayer game to play with (or against) other players, so make them do so! Make it easy. Don't force them to sit around town for hours to get a group. Don't make it a solo game until they get to the "endgame". Get them into groups right from the start…. let them auto-group whenever they're near each other. Let them help each other finish quests they share. Make it beneficial to group, so they won't WANT to solo unless nobody else is around.

Finally, give them some control over their destiny. Make something for players to achive that seems to matter. I don't know what, that's up to you. Maybe they should be able to "control" land and squabble over whose guild/nation/whatever is bigger, or has control of resources. Maybe they can build houses or cities. But give them something that seems permenant… so if they show their friends, they can point to something on the map and say "I helped build that" or "I used to own that".

If you make a game with that kind of thinking, I suspect you'll find people want to PLAY it, rather than trying to bot their way through it. I also suspect it won't be very easy to bot, since when the bot makes a mistake, it loses a lot of progress, and it will screw up far more than a careful (paranoid) player would.
06 Mar, 2012, Jimorie wrote in the 52nd comment:
Votes: 0
Rename this thread…?
06 Mar, 2012, Zack wrote in the 53rd comment:
Votes: 0
quixadhal said:
Instead of asking yourself "How do I keep people from botting my game", maybe you should ask "Why do people feel the need to bot my game".

Not sure about everquest, although I believe it had bots but…
Diablo II had a botting problem.
World of Warcraft had a big botting problem. Back when I played you could go into some zones and 90% of the active players would be bots.

Eve online has a huge botting problem. It is widely believed that several of the most powerful player empires were built on the backs of large botting networks, often in the most dangerous depths of null sec space. (We can talk about how this is possible if you like.) Back in 2009 for example CCP banned over 6200 accounts in one day in their 'unholy rage' operation and the drop in concurrent users showed that ~15% of them vanished. 15% of their concurrent users -that they caught- were bots.

This suggests to me that mechanics like inherent danger, requiring paranoid behavior, probably aren't very good at discouraging bots because bots are very good at repetitive mechanics and noticing things. A bored player can sometimes be surprised, a bot cannot unless your mechanics make it so easy to surprise things that gankers will wipe out your player population.
06 Mar, 2012, KaVir wrote in the 54th comment:
Votes: 0
Deimos said:
@KaVir: WRT your proposed questing idea, it would really depend entirely on what a quest is comprised of on your game. I know plenty of games where quests are touted as "unique", while actually being nothing more than permutations of extremely bottable components/phases. For example, "retrieve X item and deliver it to npc Y". You can vary X and Y, but a bot doesn't care. Programming in the locations (if they're static) or methods for locating the items and NPCs can be trivial in a lot of cases.

Yes it definitely depends on the implementation. But repeatable components are usually used for repeatable quests, and I'm specifically talking about non-repeatable quests.

An example of repeatable components would be the quest system in the original God Wars mud; you're given a list of 3 items, randomly selected from a list of about 150 different items, and you need to collect them. This is obviously highly bottable - all you need to do is program the locations of the ~150 items on the list.

But if you removed the repeatable component, and said that each player would only ever be asked to collect each item once, then it becomes rather pointless to bot, because you'd still have to find the item before you could program its location, and you wouldn't need to go back again. The only value in botting it would be for another character who hadn't yet completed that quest, but even then it would be a finite source of reward - there would be no benefit in botting 24/7.

Due to the combat theme of my game, my quests are combat based. They usually consist of finding the entrance to a location and completing some sort of combat-oriented objective, using clues from the quest description. For example one quest requires you to climb a tower and kill the magus - but he's protected by almost unkillable golem guards, which you're expected to either avoid or disable rather than fight directly. Another requires you to defeat the Nemean lion, which is invulnerable to direct physical damage. Another requires you to enter a harpy nest and destroy the eggs - but you're constantly attacked by waves of harpies while you do so, and if you try to fight them rather than focusing on the task you'll be overwhelmed. Another requires you to defeat a powerful undead king, but when you awaken him from his sarcophagus his guards will rise up and attack you as well, so you need to eliminate them first. Another requires you to defeat a dryad, who is practically unkillable when near her tree, so you need to lure her away first. Other tasks involve defeating ghosts and elementals with specific weaknesses, using the terrain to your advantage, protecting one mob from another while destroying scenary, or simply defeating one or more opponents using specialised tactics.

Players often require several attempts to complete each task, and frequently need to adjust their character builds between attempts. It's simply not something you could bot generically, you'd need to specifically script each individual task - and you only get the reward once per task, so by the time you'd found a solution you'd already have completed it. Even sharing your bot with other people wouldn't be very effective, as your solution might require specialised powers or equipment that another character lacks. Players would need to adjust the bot for their own character, which would take more time than just doing the task manually.

Deimos said:
And coming up with truly unique and impractically bottable quests is a really tough job IMO. Not to mention an ongoing endeavor. I think it would almost be impractical, and near impossible if quests are the sole method of advancement like you alluded to.

I don't think so, although like most things it depends how much effort you want to put into it. To give a very simple example, you could go through your world database and pick out a hundred hard-to-find objects, a hundred challenging mobs, and a hundred widely spread-out locations, and make each the basis of a one-shot quest. Players can't bot them until they've completed them manually, at which point there's no incentive to bot them any more. Add some cosmetic fluff to flesh out the encounters, and provide a list of objectives for the player to follow, and you've got the basic framework for a non-repeatable quest system.

quixadhal said:
Instead of asking yourself "How do I keep people from botting my game", maybe you should ask "Why do people feel the need to bot my game".

I already covered that on the first page:

KaVir said:
Usually for the same reason they cheat, grind, and exploit bugs, even if means missing out on the meat of game. They want to have the highest level char....

Then there are the players who actually enjoy writing bots, and view it as a sort of tactical minigame. But that's really a matter of policy; should you allow players to play the mud their way, or should you force them to play the mud the way you intended it to be played? Is unintended game design really so bad, as long as the players are having fun?
06 Mar, 2012, Deimos wrote in the 55th comment:
Votes: 0
@quix: I'll echo Zack; botting was a huge problem that all but destroyed Diablo 2 (and other successful MMOs) for legit players. And I don't think the relative lack of botting when compared to MUDs has much, if anything, to do with game design. I think it's mostly a result of how easily bottable the medium is. I can create terribly advanced and efficient bots for just about any MUD, but ask me how to do so for a graphical game of any sort and I'm completely lost.

I do agree with your comment about needing to make game design decisions to mitigate the desire to bot, though.

@KaVir: I think you underestimate the usefulness of bots for things a character can only do once. If I anticipated making more than one character ever on your game, or had friends who could benefit from it, creating a bot is still immensely useful to me as long as the tasks I'll be automating are undesirable for me. Once the bot is created, it can be shared, and suddenly tens of players are advancing in your game with no manual effort.

In short, I don't think the uniqueness of quests really reduces the utility of botting much. Not to mention that we haven't really even touched on other bottable systems like skill improvement, crafting, etc. which can't be addressed through "uniqueness."
06 Mar, 2012, KaVir wrote in the 56th comment:
Votes: 0
Deimos said:
@KaVir: I think you underestimate the usefulness of bots for things a character can only do once. If I anticipated making more than one character ever on your game, or had friends who could benefit from it, creating a bot is still immensely useful to me as long as the tasks I'll be automating are undesirable for me. Once the bot is created, it can be shared, and suddenly tens of players are advancing in your game with no manual effort.

I already covered that:

KaVir said:
Even sharing your bot with other people wouldn't be very effective, as your solution might require specialised powers or equipment that another character lacks. Players would need to adjust the bot for their own character, which would take more time than just doing the task manually.

Bare in mind that what I've been discussing here isn't simply baseless postulation and theoretical guesswork, but real solutions I've gradually introduced, experimented with and refined, in several different muds, over the last 17 years. Some solutions have proven more effective than others, the most effective ones have come at a price that I'm not willing to pay, and there is no "silver bullet". But there are certainly a number of ways to significantly reduce the value of botting, while still providing a fun game for the players, and I firmly believe that's worthwhile goal to strive for.

Of course YMMV, and if you'd rather spend your time on other parts of the design then there's nothing wrong with saying so - we all have limited resources, after all. Most muds follow the solution Runter outlined, and they seem to do well enough:

Runter said:
Or don't do that, and make a game where you just type kick over and over and tell people they can't automate it without risk of staff intervention. Which ever one seems more reasonable. :)
06 Mar, 2012, quixadhal wrote in the 57th comment:
Votes: 0
Deimos said:
@quix: I'll echo Zack; botting was a huge problem that all but destroyed Diablo 2 (and other successful MMOs) for legit players.


But that's the thing. It didn't. Diablo 1 was destroyed by hacks and cheats, Diablo 2 – 10 years later – is still going strong. Sure, it's impossible to use battle.net chat due to all the spam bots, and it's also very likely that if you play all day long you'll see bots sneak into your open games and leech things… but it's still quite playable by people who want nothing to do with bots.

This comes back to one of the things I tried to touch on in my novel up there. People like to play games with other people, and most especially with people they know from outside the game. Bots don't impact that, unless you're in a free-for-all PvP environment. If I and and handful of my friends decide to roll up new characters and play some game twice a week, the worst a bot can do is steal a boss that we were going to go kill. Other players do that too, which is why instancing is so popular nowadays.

To touch on Zack's mention of EVE's bot issue. I would submit that in that case, it's an indication of two seperate issues that overlap in that particular case. EVE itself is easily bottable, because harvesting operations are extremely repetitive, and combat outcome is affected as much by your build strategy as your playing skills. That, by itself, isn't an issue. There are tons of anti-bot corporations that hunt down macro miners and other forms of bots for amusement. Once you identify a bot, it's pretty easy to figure out how it responds and find a way to kill it quickly whenever you see it again.

What made it an interesting thing is the idea of an entire corporation (guild) deciding to bot en-masse. Individual bots are not a threat to anyone. Small wolfpack groups are, but are also easily destroyed by groups of players. An entire force of hundreds of bots, coded to work together… that's tough.

But, ultimately, the reason the bots were an issue was that activities in the game were not FUN. Some people enjoy watching TV while their mining laser bores into asteroids, giving it only enough attention to switch targets when they run out, or warp the orca in and out to push ore to station. Others can't stand the idea, but still want the wealth that comes from doing it, so they write bots to do it for them, while they play in another window (or do something else).

Also, the issue of gold sellers creeps in here. Because people will pay real-world money to avoid tedium, or to get an unfair advantage, there's an industry in automating the collection of resources to sell. CCP cut that problem down in a fairly unique way, by allowing you to purchase game time as tradable in-game items, which can be consumed for time OR sold to other players on the in-game market. That means, smart people will just buy those from CCP themselves, and sell them for the in-game stuff or money they want, cutting the gold sellers out (and thus avoiding the risk of being banned).

To solve the problem entirely though, you need to make those collection activities interesting enough for people to want to do them, and dynamic enough to make botting them annoying. Of course, that's hard. How do you generate content that is intersting AND non-repetitive AND not easy to predict? Players will always consume content far faster than you can make it by hand, so you can't just rely on hiring more people. If you solve that problem, you may have the next WoW-killer. :)
06 Mar, 2012, KaVir wrote in the 58th comment:
Votes: 0
quixadhal said:
How do you generate content that is intersting AND non-repetitive AND not easy to predict? Players will always consume content far faster than you can make it by hand, so you can't just rely on hiring more people.

In terms of botting, it's doesn't matter whether or not the content is interesting - people will still bot it, if they feel it gives them any sort of advantage.

Far more important is making the content non-repeatable. And you're dead right about the creation-to-consumption ratio, but that's not necessarily a deal-breaker, because you only need to worry about content that's used for advancement.

Imagine a mud in which it took 1 week to create enough content to provide 1 hour of entertainment. Obviously you're going to lose the content race, at that rate you'll never keep up with the players - but you don't have to. If you spend 50 weeks developing some really cool non-repeatable content, then you've got a game where players can reach the top level in 50 hours. Then you can add repeatable and sandbox end-game content. If players decide to bot the end-game content, and it doesn't give them any advantage over the other players, then does it really matter?
06 Mar, 2012, Rarva.Riendf wrote in the 59th comment:
Votes: 0
David Haley said:
I think it's great that your game has this or that property that has this or that consequence, but this is a discussion about games in general, not only your game. It's not very useful to reply to a general comment that says "botting can be really useful when XYZ is true" by saying "well in my game XYZ isn't true so botting isn't useful". You're not having the same conversation and it just makes things really confusing and it feels like disagreements for no reason and nobody is learning anything. :sad:

Hum I thought this conversation was about bots were best in EVERY (situation (for Deimos), and that you had a need to prevent botting.
As most people, I feel that preventing bots is -mostly- impossible, and that in the end it harms the honest player more than naything.
And that a best goal to have is to make so bots are not that useful in the first place. Hence MY solution to ONE problem. (I do notcare about people botting against mobs, till they are reactive enough when you talk to them, kinda agree with Hades Kane on this one).
06 Mar, 2012, Deimos wrote in the 60th comment:
Votes: 0
@Rarva: I never said bots were superior in EVERY situation. In fact, I disclaimed that very notion several times in my posts. I'd paste the quotes but I post from a phone, and that's a PITA.
40.0/72