[tz] Proposed reversions, for moving forward

Paul Eggert eggert at cs.ucla.edu
Thu Aug 7 16:40:26 UTC 2014


Derick Rethans wrote:
> I realize it's because acronyms changed *only* here

Yes, and although it was merely a regression test, it helps bring better 
perspective to the recent discussion about removing questionable old 
data.  I've long expected that 2014f's acronym changes will cause the 
most disruption to end users, that the Russia changes will be noticeable 
but not that painful, and that the removal of questionable old data will 
cause no significant problems in practice.

We thought the old acronym entries were not right -- even though we 
couldn't *prove* this -- and so we improved the data as best we could. 
Although we didn't make the changes lightly, we valued correctness over 
stability even when we knew we didn't achieve 100% correctness.  This 
has long been common practice in tz maintenance.


More information about the tz mailing list