On 29 August 2013 21:56, Paul Eggert <eggert@cs.ucla.edu> wrote:
On 08/29/2013 01:06 PM, Paul Goyette wrote:
The new proposal (appears to) restore the recent status-quo I'm afraid that's not the case: the proposal would impose new political rules that have never been followed in the tz database, rules that would require politically-inspired makework. For example, it would forbid the longstanding link from Europe/Rome to Europe/Vatican. There is no technical reason to forbid that link; the only reason to forbid it would be political. (This is not the only example; I'm picking on the Vatican because it's relatively uncontroversial.)
The Vatican and Rome may have exactly the same time-zone rules but they do not have exactly the same LMT. Now I'll accept that the discrepancy is likely to be irrelevant to pretty much everybody, but it is a necessary effect of the two rules. The real aim of rule #2 is to stop things which are politically charged like Belgrade vs Zagreb. It also handles things like Aruba/Anguilla/PuetoRico which may not be politically charged now but could be in the future. If it were to ease things for others, I would be happy with rule #2 being should rather than must, although I personally feel its a step backward. Similarly, I would be OK with a city (zone ID) linking to two different ISO3166 regions if time is and always has been the same in both parts of the disputed city - I just think that scenario won't actually ever occur. The two rules do, unquestionably, make zone IDs a function of regions. I think that is a good thing. Note that I said zone ID within region, not city within region. This distinction is where the separation from politics can occur for the few that complain. In addition, the number of controversial IDs is very, very small. They can and should be sidestepped via an "avoid controversial zone ID names" rule - ie. use Tel Aviv, not Jerusalem. Specifically, no one can complain that Tel Aviv is in Israel and that Israel has an ISO3166 code. Politics sorted. Stephen