On Feb 13, 2008 5:40 PM, David Gerard dgerard@gmail.com wrote:
On 13/02/2008, Cool Hand Luke failure.to.communicate@gmail.com wrote:
As for the evidence, 27 and counting disagree with you, and the others
who
are dragging their feet even though they have long suspected a COI.
Yes, a torch and pitchfork count is definitely conclusive evidence. Well done.
David, did you read the evidence? Nothing's ever conclusive in these issues but I, who have never edited these things or run into these people before, think that it would be difficult to file a more comprehensive indictment of a sock, and definitely this is the most efficient I remember. (There's always the probability that we're wrong, but the methods in this case have reduced it to considerably less than it has been before, and that probability never seemed to have bothered us much in this area, did it?)
GWH, I'm all for ensuring that similarities are checked with the appropriate control set, but the bit on "interleaving" in the timestamp statistics comparison is pretty damn convincing, and independent of time- or area- specific control sets.
RR