G. wrote: > You know, I haven't followed all the ins and outs of the arguments, but > I really don't thing dependent actions have to be judged broken. > > Consider: > > Originally, the rule relied on a single list-based linguistic convention > to decide whether the A, B, and C were logically 'A and B and C' vs. > 'A or B or C'. (The IFF contains no qualifier like "IFF all are true; > if at least one is true"). The original rule was IFF A ,B and, C which, > in linguistic convention, maps to the logical IFF A and B and C. > > The new rule is IFF A, B and, C, D which breaks the common linguistic > mapping. So there's no absolute guarantee that this maps to IFF A and > B and C and D. Given the placement of the 'and', it could just as easily > map to IFF (A and B and C) or D which is not broken at all, reasonable, > and in keeping with the intent where the rule is unclear and for the > good of the game. The alternate that A, B and C, D => A and B and C and D > does not keep to linguistic convention, is against the intent, and not > particularly for the good of the game. There's no support that one > should infer "IFF all of" versus "at least one of" in list qualifiers > in general. > > Forgive me if this has been refuted already, I may have missed a subtlety > and this is up to a judge (given an actual judgement), but, just sayin'.
This is basically how c. argued when judging CFJ 2670.
