August 21, 2005

Computer characters mugged in virtual crime spree

I often say that you can only tell if some FC application is successful when you see crime start to appear. Games are now massive successes in creating value and payment systems, again showing that the Finance layer is more important than the rest put together.

Computer characters mugged in virtual crime spree 11:31 18 August 2005 news service Will Knight

A man has been arrested in Japan on suspicion carrying out a virtual mugging spree by using software "bots" to beat up and rob characters in the online computer game Lineage II. The stolen virtual possessions were then exchanged for real cash.

The Chinese exchange student was arrested by police in Kagawa prefecture, southern Japan, the Mainichi Daily News reports.

Several players had their characters beaten and robbed of valuable virtual objects, which could have included the Earring of Wisdom or the Shield of Nightmare. The items were then fenced through a Japanese auction website, according to NCsoft, which makes Lineage II. The assailant was a character controlled by a software bot, rather than a human player, making it unbeatable.

Ren Reynolds, a UK-based computer games consultant and an editor of the gaming research site Terra Nova, says the case highlights the problem of bots in virtual worlds. Arms race

By performing tasks within a game repetitively or very quickly, bots can easily outplay human-controlled characters, giving unscrupulous players an unfair advantage. Many games firms employ countermeasures to detect this bot activity. For example, they can ask the character questions or present them with an unfamiliar situation and monitor their response.

"There's an ongoing war between people who make bots and games companies," he told New Scientist. "And making real money out of virtual worlds is getting bigger."


Posted by iang at August 21, 2005 12:06 PM | TrackBack

I don't think that this is "unfair" advantage. If anything, it's a flaw in the game's design.
There was one a quiz game on Hungary's largest daily's website, where one had to answer 30 questions as quickly as possible, and there were some prizes handed out to the best.
One could play the game as many times as one wished, but it was always the latest result that counted.
The game had three fatal flaws:
1. The time was measured by the player's own computer (in javascript), and taken by the server on faith.
2. Secondly, after each answer (one had to pick the right answer from three possibilities), the computer told you whether or not the answer was correct.
3. Until you answered all 30 questions, you could restart the game, without affecting your last result.

This meant that it was trivially easy to assemble a database of the approx. 1500 questions that they had, together with the correct answers. And write a code that aced the quiz.

It was very funny to watch the organizers bounce around in panic trying to patch the gaping holes. Along the way, to save face, they made all sorts of bluffs (like the game had thousands of questions, etc.), that were, of course, immediately called. Finally, they kicked out a lot of players based on a visual inspection of the logs and then held a final round in their head office on "tamperproof" computers to select the winners. And most importantly scrapped the promise to play such games every month. This was the first and the last one.

Posted by: Daniel A. Nagy at August 21, 2005 03:38 PM

in 1980, the person responsible for REX(X) had done a mainframe 3270 multi-user spacewar game. basically somebody started a spacewar game server ("the umpire") and local users (on the same time-sharing machine) and remote users could all connect and play against each other.

there was a provided client ... but the protocol interfacing to the server was fairly straight-forward and eventually somebody wrote their own automated bot and started beating everybody.

the game was then modified to consume energy inversely proportional to the time beween client commands ... below some interval threshold ... the amount of energy used started going up dramatically. this helped to level the play field between the automated bot players and the human players.

aka supposedly the big advantage of bots over humans wasn't so much better strategy ... but being able to play their moves faster ... so drastically increasing energy consumption for fast players ... introduced some additional player strategy trading off speed of play against energy consumed.

lots of past spacewar posts IBM 1130 (was Re: IBM 7090--used for business or science?) oddly portable machines A question for you old guys -- IBM 1130 information Z/90, S/390, 370/ESA (slightly off topic) 5-player Spacewar? 5-player Spacewar? 5-player Spacewar? 5-player Spacewar? Logo (was Re: 5-player Spacewar?) VM: checking some myths. Help needed on conversion from VM to OS390 Amiga Rexx 6600 Console was Re: CDC6600 - just how powerful a machine was Computer Terminal Design Over the Years PLX Vector display systems Wanted: Weird Programming Language Re : OT: One for the historians - 360/91 OT: One for the historians - 360/91 Why only 24 bits on S/360? The PDP-1 - games machine? 1130 Games WAS Re: Any DEC 340 Display System Doco ? Any DEC 340 Display System Doco ? instant messaging Seven of Nine IS CP/M an OS? A POX on you, Dennis Ritchie!!! Playing games in mainframe who were the original fortran installations? Usenet invented 30 years ago by a Swede? Whatever happened to IBM's VM PC software? RISCs too close to hardware? Where should the type information be?

Posted by: Lynn Wheeler at August 21, 2005 04:09 PM
Post a comment

Remember personal info?

Hit preview to see your comment as it would be displayed.