Non-free anti-cheat libraries
Mon, 2007-10-01 18:59
As the good guys at BZFlag IRC channel warned, some games apparently have anti-cheat methods built in as non-free libraries. This is a rather peculiar issue considering the purpose of anti-cheat mechanisms, to prevent cheating, which is why there is some sense in hiding the source code of such mechanisms.
So are we to blanket accept such things in our games for this tourney? What do you think?
I'm not sure yet which games exactly have this, but it doesn't seem to be talked about a lot, which is why I initially missed out on this issue. I suppose most people do take this for granted as a non-issue.
Anyway, thanks for your feedback!
Danijel
Non-free code is unacceptable, no matter what the purpose is. If they want to prevent client-side cheating, they should bloody well properly enforce it with server-side checks and whatnot. (Though these may be impractical for performance reasons...)
Alright, here is the thing. I tend to agree with that free-zombie, so we should find out whether any of the considered games has such code. However the larger issue is preventing cheating, which is being strongly touted with regards to BZFlag. Apparently there are a lot of cheaters among BZFlag games and this could become a serious issue.
We don't want to get someone winning because they are cheating!
I think it would be nice for Cluenet guys to have a say on this and enlighten us as to how they are dealing with this issue.
I'm aware that BZFlag is easily cheated. Our very own tonyb coded his own GL-hook that made him able to see through walls. Loading simple hooks like this is easily done with LD_PRELOAD=/path/to/lib bzflag, iirc.
If it's between us, we know that none of us use cheats, but if we get some new public users, we may get some raging myg0t-wannabies.
It's a fuzzy area indeed.
practical argument against anti-cheat-blobs: They lock out everyone who doesn't have or use $platform. That could include PowerPC users, BSD users, or x86_64 users that don't or can't use x86 blobs for some reason (e.g. if their GL libraries are x86_64 only or something...)
Well, it is not just between us. We already have more than 20 players signed up and most aren't just us. After the two videos and the final announcement is released, and especially if it gets to digg, it is quite likely that we will get even more public users so I think it would be good to have some sort of a strategy about this.
Not sure what you're getting at. Could you reword that?
For non-free anti-cheat libraries to be useful, they must be binary blobs to make cheating as difficult as possible. This means that the game developers release binary code. As they are unlikely to release blobs for all platforms the game could theoretically run on, that artificially limits the player base, locking people whose systems aren't explicitly supported out.
If you needed a Linux/x86 blob to run Nexuiz, Linux/PPC or NetBSD/teapot users couldn't play it.
?
What's a teapot? New architecture?
ah. fair enough. and what's teapot?
Tonyb said he might make a blob where all source would be free and available except a special key which can be changed for each game, if I got that right.
In any case since the new key can be generated always for a new game it should be a way to have an anti-cheat blob without technically being non-free and having these disadvantages.
Some encryption-based DRM could be portable and of doubtful freeness.... "NetBSD/teapot" was a joke.