What nonsense.. KC-KC-Outpost-Masquerade is FAR crueller than a goons pin. It trashes 5 cards/turn AND is resistant to any form of reaction card (Moat, horse traders)
Let's do some threat modelling:
1) I could cheat by lying to the server about what cards I have in my deck or in play when I play a card or buy a card.
For buying: Fixed by DougZ.
It's possible there are other versions of this exploit, where I could tell the server I want to play a Witch, but I don't actually have one. Any instances would be easily caught by someone paying attention and storing the isotropic or council room log.
Probability that exploit exists: Extremely low
Visibility of exploit: extremely high
Utility of exploit: extremely high
Not worth worrying about. If it's happening, someone will post a log.
2) If the client-server protocol stores a copy of the game state on the client, I could cheat by examining the protocol, thus knowing what cards are in my opponents hand, and the order of cards in both decks. This would help in cases where I'm considering breaking the PPR, or if I have to decide whether to play e.g. militia or wharf with my last action.
Probability that exploit exists: unknown, but likely low*
Visibility of exploit: low
Utility of exploit: medium
* I say low because there is no reason to do it this way. Also, at times, there is a visible lag when playing a card that draws cards, which mostly likely means that the client needs to do a round trip to the server to figure out which cards to draw.
If there is any non-visible game state on the client, this is the most likely way a player could cheat.
3) Player could use an advanced AI to cheat. Poker players in particular are likely extremely sensitive to this possibility, especially if they haven't spent a lot of time playing Dominion. Most known AIs poorly handle many complex situations. Also, Dominion is extremely dependent on the shuffle(to a greater extent than poker), so it's harder to calculate the right move. You could build an engine that would tell you things like "odds of drawing a dead action if I play this smithy with my last action," which cards are in the discard/draw deck, tracking of money, etc.
Probability that exploit exists: medium
Visibility of exploit: low
Utility of exploit: medium
Not much you can do about this one. As a mitigation, there is nothing at stake in isotropic, so what is the motivation for building such an engine? We also know that most existing AIs are terrible.
4) Player could prepare a simulator to help decide the opening plays. On simple boards(especially where BM+X is the dominant strategy), this could help a player start with an advantage. Not as useful with complex or engine boards. Also, the strength of most cards is pretty well known, so this is less likely to help at high levels.
Probability that exploit exists: known to exist
Visibility of exploit: medium(delay at start of match)
Utility of exploit: low-medium(on some boards, can provide an advantage)
Also not much you can do about this. I think this will turn into something like a chess opening book. Strong players will have to learn it, and everyone will use computers to aid their analysis.
Can anyone think of any others?
"Trash question" seems like a awfully scruby thing.
Also, this.
Does it have to do with what the card does?
If I am right this is really easy....
University?
