Zoidsoft wrote:
so they probably did an interpolation and preferred weekends for setting this change
That's pretty much what I inferred from looking at the Shanks data. Partly it's because I use the same process when doing my own research. See, for example, the latest change to Pacific/Johnston. For the proposed changes to America/Anguilla etc. the most likely explanation is that Shanks et al. lacked transition info and guessed, and (worse) guessed different transition times in different locations because they made individual guesses and didn't check their work later. The tz data have long been maintained trying to fix stuff like this when I find it. In several places I've corrected Shanks data when I think it's bogus, even if I can't prove it's bogus. These are noted in the comments.
they guard this information as to exactly what criteria they used as a secret
That's where we differ from Shanks. The new Pacific/Johnston entry has a comment saying that it's guesswork, and giving the primary source on which the guesswork is based. It's another place where Shanks is incorrect and I'm supplying my own guesswork, but at least I'm trying to note my assumptions. Changes like this are a longstanding part of maintenance, and I'm becoming inclined to think that we shouldn't discontinue this practice purely from a desire to not change things.