Vorsch wrote on 07/14/11 at 03:01:49:On smk, there is Guillaume Bertrand, which is a much more dangerous kind of cheater than Fenner ever was. He cheated multiple times on multiple games. He also used fake screenshot / videos to support his claims, which, as far as I know, is something that Terrence never did. On mk64, the members of the Hab trio (three people that reached #1 under a false name) are still allowed on the site, and even have 4 WR (three untied). The list goes on and on...
Hab trio is a completely different situation, as they never submitted actual fake times to the site. Though if it were up to me, they would be banned as well yes. People have to learn to play by the effing rules.
As for Guillaume Bertrand, could you specify which games he was caught cheating in? And how this happened? He originally tried to pass his TAS-ed SMK runs as WR's, but never submitted them as such to our site, so technically he hasn't attempted to cheat. I agree, it's a shady case though. He should probably be removed as well, in the strictest sense. I can't think of others on the SMK site; the list certainly doesn't go on and on. Also, I'm fairly certain Fenner has been caught cheating at least twice as much as the next most caught cheater in the MK scene.
Vorsch wrote on 07/14/11 at 03:01:49:The goal of a proof policy is not to verify that a member is trustable. Why? Because such a thing is impossible to verify.
What can automatically be derived from your reasoning is that there is no proof policy known to man that can resolve the Fenner case. Like you said, you can only determine to a large degree of certainty, which times are real and which ones aren't. But that's not the most important thing.
Vorsch wrote on 07/14/11 at 03:01:49:That's why a proof policy based on trust can't work.
Trust is the only sustainable way through which these websites can be maintained. You cannot possibly verify every single time of every single member, therefore we have to make the assumption that we can trust a member. Proof policies come into play once certain performances come under suspicion for whatever reason. Once these policies have caught a cheater, it means that person has breached the trust system, and has thereby demonstrated a complete lack of respect of the ultimate cornerstone that these rankings are based on. You have to understand that proof policies are not meant to 100% verify every single performance of a karter from a certain point onwards; that'd be quite ridiculous and the amount of effort that is necessary for such a high sustainment of proof is completely out of proportion. No, the goal of the proof policy is to demonstrate if someone is cheating; if not that person can be incorporated into the trust system again (eventually).
As Florent stated, once you have proven that a person can not be trusted (that's more often than not the way science works; you can't prove a positive, you can only eliminate which scenarios are false), that should be it. Especially if it is proven reproducably.
It boils down to how you view the site. Is it a place where you just come to fight against performances of other karters (your only interest are the times), or is it a place where you can enjoy the competition with people that respect the community and your mutual trust?