Paul Eggert wrote:
Is the March 2016 announcement the first time a government has cited tzdb,
Watch out for the next milestone in this direction: the tail wagging the dog. Look what happened to country-code TLDs. It was a great idea (of Postel, I think) to delegate the question of what is a country to ISO 3166, and this worked fine for the first decade or so. But when the Internet reached the masses, ownership of TLDs became so significant that it provided a motivation to manipulate ISO 3166. The original rationale for using it, that it's a neutral arbiter and in particular unmoved by what goes on on the Internet, was lost. How would this manifest among timezones? If tzdb is taken as the arbiter of what is a timezone, which it indeed functions as for some purposes, maybe there'll be a desire for each country to have its own (non-link) timezone, as a matter of national pride. Countries that have had identical time since 1970 might diverge for a few months just to get themselves (or rather their capital cities) on the list. Sub-national regions with separatist ambitions (e.g., Catalonia) and enough devolution to achieve it might pull the same trick to get their own timezone, so that they can say they have the same status as sovereign countries. For this to start happening just needs governments to perceive that appearing on computers' lists of timezones is desirable. The payoff isn't as obvious as the benefit of controlling a TLD, and it all requires a much greater popular awareness of computer timezone handling than has previously been the case. But the Azerbaijani government citing tzdb shows that we're well along an upward trend of the latter. I think the whole scenario is plausible, for a few years from now. -zefram