Eve Bots - are they a bad thing, is CCP complicit in their use?

CCP cant do it that way for legal reasons.
That isnt enough evidence.

Sufficient for internal flagging for review, perhaps, but not enough for an account ban.

Also, not all bots run all day, or everyday of the week.
Some also move around.

Im not going to go into specifics, but detecting bots is not as easy as you might think.

There are extreme options such as CCP logging and crunching the enormous amount of data sent by player activity, or outsourcing crunching of that data to a 3rd party, but those also carry some pretty severe/questionable legal issues.

I think CCP also has problems with how data is sent/received. Not all bots “spam” commands, some issue them once in the ~1s delay allowed for in the EVE system. This makes it difficult to differentiate between a human playing really hard (whom sometimes also spam commands), and a bot. This is barely scratching the surface of this and other complications.


Hence, as I proposed above, implement a system that prevents rudimentary bots from operating in the first place, which should comprise most of the bot volume.

A simple captcha, of some form, will wipe out most bots.

After that, CCP can focus on the remainder of more advanced bot systems, or those to which the captcha system cannot be feasibly applied.

The price for that we pay as players, is having to make a few extra clicks.
We make hundreds if not thousands of clicks per play session anyways.
Whats a few more.

Im ready to do that. I think most of you are too, if it means we cripple botting in this game we all love.

Detecting and fighting bots specifically is fruitless … the only real counter IMO is to change ISK faucets so, that AFK/scripted play expose the players to a such high risk, that bots are not profitable anymore.

My long time favorite are still tags instead of bounties, combined with restrictions on MTU usage (so bots can’t just sit on it and loot every second). Give players an incentive to profit from hunting ratters and bots.

Other means are resource depletion, so you have to switch systems or pause, and responses to the size of vessel you bring … bring a dread, spawns an equal counter, which points you and you have to kill first.

1 Like

Agreed.

No single player sat at a kb could possibly log in for that amount of time day after day, week after week, month after month.

Those accounts should be suspended until checked thoroughly for botting or account sharing, then given a permaban for botting and a temp ban as a warning for the acc sharing

This does nothing against mining bots or certain mission type bots.

Sounds like a stealth “Ban AFK Cloaking” idea to me.

No, but resource depletion and proper rat responses can address mining bots. No PvP backup, you die.

Yeah I’m not suggesting immediate ban, just flagging for review. As to your other point, I’m not aware of much legal recourse for account bans. The EULA gives them a lot of latitude on that.

1 Like

No.

Mining bots can be scripted to use drones/modules for defense.
They can also tank to the gills, since they dont care how long it takes to mine the field.
Resource depletion is where the botting player has to find a new field, manually, anyways.
Only extremely advanced bots, if any, will be able to automatically detect a field is empty and move to find another one that isnt, and restart mining, without player input.


Changing ratting bounties to tags will just result in loot/scavenger bots flying alongside them.


The solution is to systemically prevent botting, via means of a simple captcha most bots cant overcome.

The key is forcing human input.

Nope, tbh it never crossed my mind.

I’m with you, but I would like to have an EvE way and not some artificial checks. We can figure out what activities are AI hard to overcome, maybe finding a new resource is one, but also PvP is. Today bots just warp to safety if something non-blue appears in local - take this away, and see what happens.

The EVE way, would be us as players detecting and wrecking botters.
We can still do that, post-change.

The rational, and ironic way, however, is responding to artificial play with an artificial barrier. Bots have a unique kind of stupid, that no player has.

If we are going to fight bots in our midst as essentially AI systems, it makes poetic sense that the best way is for us to prove our humanity, which they cannot emulate.

PS: Bladerunner/“Do Androids Dream of Electric Sheep”

I’m not an AI/bot expert, but I do know more than most.
Ive looked into bots because I hate bots, absolutely.
I want to know my enemy.

A simple 3x3 grid pop-up with 2 dots of a randomly selected color that you should click to connect them should be enough to completely wreck any but the most advanced bots.

The colors can be chosen so as to not impair the color-blind.

Thats 2 clicks for attending, present players, whereas even scripted AIs programmed to click random fields in the 3x3 grid will likely have to run through hundreds of attempts on average before success. I’m too math stupid to calculate the odds, but they will certainly wreck a botters day. (Can any math savy run the math on chance of randomly choosing two connected points in a 3x3 grid? + Would the odds be higher if those two fields are not adjacent/connected?)

If the 3x3 grid pop-up shows up in a random area on the screen, we can further discount all but the absolute cutting edge bots.

Increasing that to a 4x4 grid, even further confounds the statistical likelihood that an AI randomly clicking fields will ever find the two matching dots to connect, exponentially.


This captcha system will work against the huge majority of EVE bots, especially against prolonged afk botting.

There may still be exceptions, but A) CCP will have more time/resources to delve through data to find the remainder B) Those cutting edge botters will not share their tech/setup with anyone.


What remains to be discussed, is when/why this captcha pop-up should occur, so as to impair non-botters as little as possible, and botters as much as possible.

That sounds like the seller trying to soothe the client with lies. People who pay money are banned too. Not even when they bot, can be they just say something wrong to the other player. Like GigX.

I think we all know that CAPTCHA is not a solution - the web bots proved that.

I like what is google doing with re-captcha (i think it is google) - tracking mouse movement and another data on various sites to determine bot.

You are making false correlations.

I think you don’t “know” anything about what is involved, let alone “we all” knowing it.
You are jumping to false conclusions on false premise.

A pop-up in EVE, is different than a pop-up in a browser.

Most botters in EVE do not have access to software or know-how to create a bot capable of overcoming even a 3x3 grid so as to connect two similarly colored dots in that matrix.

This is a different topic and incredibly invasive of client behavior.
It leads to website structures where as the cursor from the client passes over elements in the page, it sends data to the provider.

You might “like” that, but I sure as hell dont.
In anycase, its not relevant to EVE, as EVE does not track mouse movements, only actual actions via keyboard/mouse clicks as sent to them via the client. They dont track anything else.


Captcha bypassing tech/software (by whatever means) is not something the majority of bots in EVE have access to or know how to use.

Until someone writes a Discovery-bot to discover bots. :grin:

–Gadget is in Botception

2 Likes

Correct, a CAPTCHA won’t stop a bot because the player will just do it on log-in.

Take for example the dude that lost 8 nyx bots. 8 accounts is NOT an onerous activity, even if he had to type in the creds and RSA key for each.

Truthfully, no single approach will work. Anyone can come up with an idea, and anyone else can legitimately say “but what if they…”

So, while I agree that forcing human input is an excellent start, it needs to be one of a line of defenses. It also needs to be something frequent enough that nobody will WANT to do it that many times. Sadly, that is diametrically opposed to any acceptable customer experience (oh, you want to warp? fill out this captcha!) Thus, a captcha would hope to defeat a massive army on logon, but be useless against a small swarm.

Monitoring an account’s metrics is another option, but again it’s ever-so-easy to falsify that by making a bot take “breaks”. They can even log off after a random number of hours. The more you crack down on bots, the more valuable their isk will be, so the loss of production is offset by the increase in inherent value.

Another very valid thing to do is to simply scale up the difficulty to the point that it’s so god damned hard to bot it that you won’t be able to justify doing it in anything more than a T1 ship. Consider the BR Sotiyo AI.

If a PVE site (mining sites included) behaved like those rather than set waves, with a scaled response that would necessitate different tactics, it would complicate a bot further, making it more likely to mess up.

Last but not least, guardian software. Yes you can defeat that too, just like everything else,

TLDR:

  • Captcha will beat swarms of bots (VNIs for example)
  • Metrics will beat “the slow but steady” bots
  • Complex mechanics will beat the “stupid” bots
  • Guardian software will defeat people who aren’t proficient with computers (which it IS surprising how many IT folks play Eve).

Nobody claimed you would enter the captcha only at login.
That makes no goddam sense, since you just obviously, manually, entered the game seconds before.

You have completely missed the entire point.

1 Like

Mouse tracking was introduced for some kind of UI research in the past In EVE, but I dont know if it was in office lab or in actual official client. I remember seeing some screenshots and there was a cloud of mouse movements there. May be they showed it on some fanfest. Was a long time ago.

1 Like

Mouse movement is irrelevant for purposes of EVE.
I can move my mouse in random circles, in different cycles, before each click to enact an action. That random mouse movement can be scripted into bots, or, though extreme, automated by engineering on the client side.

All that matters, is when/why the client sends data from a click for actual action in EVE.