Slashdot Mirror
Xbox Live Labels Autistic Boy "Cheater"
Jellis55 writes "Jennifer Zdenek, the mother of an 11-year-old boy who lives with autism, is outraged at Microsoft Xbox Live for labeling her son a 'cheater' and taking away everything he's earned online. She says her son, Julias Jackson, is so good at playing X-Box games, Xbox LIVE thought he cheated. She says her son got online last week to play Xbox LIVE and saw that he was labeled a cheater and had zero achievements. Microsoft continues to ignore her requests to take 'cheater' off of his account."
Maybe he actually cheated... LOL. Naturally, the mother is biased in favor of her son.
I once played an online game where you could set the robotic factories to building robotic factories, and then after a while switch them over to building ships. In one turn you could produce a huge fleet out of nowhere. When I did this, the game designers were convinced I had cheated because "there's no other way you could get that many ships." They didn't understand their own game, or how exponential growth works. Explaining this didn't help, I was banned.
P.S. So in the next round I helped my friends actually cheat by hacking the game's database and producing written spy reports of enemy movements. Ha.
So we're just supposed to believe the person who banned him without any details. Is twitter now a reliable source?
He sent the details to the parents. Those are the only people he has to convince. Whiny mouth breathers on /. demanding that MS provide proof are not on his list of people he has to convince or impress.
If you find that a legless man is hiding items in his wheelchair you can still kick him out of your store.
For the reporter to ask: "What's your autistic 11 year old doing spending all his time playing Mature rated games that revolve around killing people?"
Shift happens. Fire it up.
Who says autistic children can't cheat? Where is the evidence that supports that assumption?
Any sufficiently advanced influence is indistinguishable from control.
In this sort of case, the issue is largely the use of modified save game files. You can use standard USB storage to save games nowadays, so it shouldn't be suprising that there are tools out there to write modified saved game files which put you right at the point you get the achievements to them.
It's actually been around a while, there were adapters that let you write modified save game files direct to the hard drives of XBox 360s too.
I suspect Microsoft does some kind of signing per-console or per-player or something on files when they're written to storage, and if the user loads a save file not signed to a console they've used or their account then it's flagged up to Microsoft.
So they're not necessarily using any kind of heuristics based detection as this mother would seem to suggest, it's likely just that as they said, he actually cheated, and mummy decided to make a fuss out of it without knowing the full story.
I decided to investigate a little and found his gamertag (ZOMBIE KILL67). Looking up his stats on bungie.net for Halo 3/ODST:
http://www.bungie.net/Stats/Halo3/Default.aspx?player=ZOMBIE+KILL67&sg=0
Ranked K/D Ratio: 0.84 over 1,014 games? Not that good after all then, in fact, if he can't even break even and gets killed more than he kills, that means he's worse than most other players, and that if he got banned for being too good, so would more than half of Microsoft's other subscribers.
So it really sounds more like mummy can't cope with the idea that her son is actually fairly crap, being below average, and that he likely is in fact a cheater. A case of parent/child flaw blindness I would say.
The modern way that cheating is detected is basically simple and yes it is accurate.
Microsoft has access to the internals of the game code while you play. Here are a few examples of how MS knows if someone cheated.
All the below are a few examples are checked while the person is playing online.
1) Look for the actual cheat code in the game memory.
2) Look for the specific altered values of game code that represent what the cheats modify.
3) Look for the jump point in the game memory, that the cheat code uses to hide the cheat outside of game memory.
4) Look to see if a specific file has been altered from what is expected and/or is impossible to have that value(s)
5) look at the supporting system files the game uses for specific alterations from standard.
The methods used are accurate in detecting cheaters. I've simplified the full explanation of some and left others out. I've just listed common methods that a lot of people are aware of already. My point here is if these or similar are the methods MS used, then the kid is a cheat.
It's been my experience that parents just don't want to believe their little Johnnie or Mary is a cheat. Some will once you explain how they were caught but almost as often the parent stays in denial. Denial won't change MS's view or decision however.
This woman's threat that she will cancel her child's account is a good thing. The first and obvious is that it will remove one more cheater from the on-line game community. It will teach her child a lesson. By the parent canceling the child learns a lesson that cheating has ramifications beyond what MS did to the account. Now the child will learn that just losing achievements was a more just punishment after all and by lying to his mom jut made things worse. Now he can't play online.
In my opinion MS should have permanently banned this kid from whatever game he got caught cheating in. MS should offer one redemption and that is to buy the game a second time and he can play again unless he is caught once again cheating. If the person does it again then make the ban permanent with no further second chances.
It's been my experience that showing the cheater that there are significant consequences to their actions, does work. Some are just a little slower on the uptake than others but after a couple of bans they get the message. My experience is based on what I do and it's represented by my /. alias.
I have no idea what goes on behind the scenes at Microsoft and how they detect and handle cases of cheating. However, for quite a few years during my postgrad days and early years of employment, I was involved on the admin-side of the (PC) online gaming scene, and spent a lot of time dealing with cheaters. I have a few thoughts based on that:
A ban for just "being too good" is highly improbable, assuming that MS have even mildly competent people working on this. Back when I was running a (major, UK-based) Counter-Strike league, the kindest description of my own level of play would have been "slightly better than average". There were players in the league who could have beaten me with their eyes closed. My admin team contained people who had a range of ability levels, but none of them were top-tier players.
Adminning top-division games was therefore something that had to be taken very seriously. Accusations of cheating were always rife in CS, though in my experience the actual level of cheating, outside of a relatively small proportion of badly adminned public servers, was never as high as it was commonly perceived to be. Making sure that average players were able to tell whether a top level player was cheating or was just plain good was, therefore, one of the main challenges for an admin team and one that was taken very seriously indeed.
We had a number of principles in place regarding accusations of cheating (or independent admin suspicions when no accusation had been made). These were:
1) Any flags raised by the Valve Anti-Cheat were treated as reliable. If VAC says a player is cheating, they are kicked from the match and the league immediately. They can appeal, but would need to show very convincing evidence that there had been a false-positive (nobody ever managed this, all we ever got was "OMG my brother installed cheats").
2) Knowing that Valve Anti-Cheat was, at the time, fairly easily defeated, admins were expected to know the signs of cheating and to watch for these. We had a library of video clips that all new admins were expected to study, some of which showed players who were using wallhacks or aimbots, others which showed clips that were just of very good players pulling off shots that looked suspicious, but which were recorded at LANs and verified as legitimate.
3) If an admin suspected that a player in a game he was refereeing was cheating, he did not stop the match or kick players. No bans were given at this stage.
4) Admins recorded all matches as a matter of policy (both for anti-cheating and because players liked to download the replays later). The admin of the "dodgy" match flagged a concern in private to the senior admins.
5) 3 other admins, including at least 1 of the senior admins (usually me) scrutinised the demo from the alleged cheater's point of view. There were reliable signs of cheating (as opposed to good or lucky play) that could consistently be spotted. One classic, though by no means the only sign, was an instant-flick moment of the crosshair to an enemy, completely out of line with that player's usual mouse-sensitivity.
6) If 2 of the 3 other admins (with one of the two being the senior admin) agreed that there was cheating, then the player was banned from the league and the results of games they had played in were overturned (subject to appeal). If there was no consensus, then the original admin who raised the concern was thanked for their diligence (there was no harm in privately flagging suspicious activity - I always encouraged it) and no further action was taken.
In around 75% of cases, all 3 reviewers would agree that there had been no cheating. In around 95% of cases, 2 of the 3 agreed that there had been no cheating. We averaged around 3 player bans per season, of which 2 were usually as a result of "technical" (ie. VAC) detections. I am confident that none of the "admin" detections that were confirmed were false-positives.
My point is that this is the degree of scrutiny we applied to what was, for most of its
Adding to the GP and P posts: I also play online FPS games, I am an admin with reasonable experience, and, most importantly, [b]I've had the chance to see autistic kids gameplay[/b].
And here's the thing: before I found out a player was autistic, their manner of play raised all kinds of warning flags for me. There were spurts of uncanny abilities, they wouldn't talk to anybody, they were focusing obsessively on a limited sets of actions (run this exact route, attack at these exact points), they displayed anti-social behavior (attacking their own team) for no apparent reason. My first reaction? What a cheater/asshole combo!
Has anybody considered how their repetitive/compulsive nature alone may cause autists to deviate from the player norm? Not to mention that about 1 in 10 autists show outstanding abilities ("idiot savant" kind) and about 9 in 10 exhibit enhanced sensory perceptions.
So I find it strange that most highly-moderated comments so far have completely ignored the fact the kid is autistic and how it may have affected his gameplay. My own experience tells me that unless Microsoft knows for sure he used an actual bug or exploit, I'd take that "cheater" verdict with a BIG grain of salt. I'm fairly confident that an autistic person can trip both automated and human cheater detection. They were designed for regular people.
i ate crayons when i was a kid and now i have two braincells and the blue ones taste nicer