oshfdk > 01-08-2025, 02:23 PM
(01-08-2025, 02:09 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.(01-08-2025, 12:16 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.I assume the paper is this one: You are not allowed to view links. Register or Login to view.
[...] First of all, the number of authors is not known for 59% of the sample set, so even as an extrapolation the argument is not very good.
Actually it would be a good argument, IF the VMS has been chosen at random among the category of "encrypted books".
Say there is a box with 118 tokens, some marked "M" (multiple authors) and some marked "S" (single author). You pick one token at random from the box, but it drops into a hole before you can read it. You must guess whether it was an "M" or an "S". So you spill the contents of the box on the floor, and count what you see among the remaining 117 tokens. 73 of the tokens fall face down; 44 fall face up, and they are all "S". That is very strong statistical evidence that the one you picked was an "S" too.
Jorge_Stolfi > 01-08-2025, 02:55 PM
(01-08-2025, 02:23 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.If the fact about the number of authors was revealed via some uniform random process (or whatever this is properly called, I mean via a process that represents all outcomes equally well), then yes, your example is a good one. But what happened in reality is the assessment of the number of authors was easily done in the situations where the contents (e.g., a personal diary) lead to an immediate conclusion about the single author. So, it's more like there are tokens either marked S on one side or empty on both sides, after spilling the contents on the floor some of them show S, some of them show nothing.
Jorge_Stolfi > 01-08-2025, 03:16 PM
(01-08-2025, 08:49 AM)Yavernoxia Wrote: You are not allowed to view links. Register or Login to view.I don't see any stroke on either of the images posted. It's pareidolia,
R. Sale > 01-08-2025, 07:01 PM
ReneZ > Yesterday, 12:23 AM
Aga Tentakulus > Yesterday, 02:07 AM
Aga Tentakulus > Yesterday, 02:11 AM