On 5/28/21 6:16 PM, Arthur David Olson via tz wrote:
A philosophical note: maintaining and processing information about the oddball stuff that's been done in the past helps ensure the ability to cope if the same sort of oddball stuff is done in the future.
Yes, that's long been a stated goal of tzdb - you added a note to that effect to the README file in (let me check the repository...) 1986! Paul Ganssle followed up today with a similar comment. I think every tzdb feature used by 'backzone' also occurs in the other files, so backzone's practical contribution to exercising zic etc. is limited. That being said, we can be more systematic about using 'backzone' that way. Proposed patch attached, and installed in the development repository. If someone wants to contribute further improvements of automated testing of 'backzone' that would be nice. Realistically, though, 'backzone' will continue to be lower-quality than the other files, as we lack the human time to check and/or improve it despite its known problems. By the way, one problem I have with "make check_public" is its large amount of not-that-useful 'zic -v' chatter like 'warning: "europe", line 3679: rule goes past start/end of month; will not work with pre-2004 versions of zic (rule from "europe", line 3660)' which causes me (and I assume everyone else) to ignore the chatter. Perhaps we should add an option to zic controlling diagnostic age - pre-2004 is pretty old nowadays - and change the default for zic -v to not warn about stuff that's been in zic for five years or so. After all, there's no point having tests if people always ignore test results. On 6/1/21 11:42 AM, Paul Ganssle via tz wrote:
The conversation instead goes, "Here's the rule we're using" and the follow up is, "That's a stupid rule, you should change it." then a bunch of people pile on in both sides and no time is saved.
Yes, we've had our share of pile-ons. There is a distinction, though, between tz mailing-list politics (the focus of much of the recent discussion) and real-world politics (things like, "is Kosovo a country?"). My main worry is the latter not the former, in that I think it's worth making minor technical changes to tzdb now to help forestall potentially major real-world political problems down the road, problems that could be worse than being sued by astrologers. Admittedly not everyone sees things this way.