Keep in mind that if 'local convention' is 6:00am is synchronized to sunrise then the time changes every day since sunrise time is different every day (usually by about a minute or 2 but that depends of the time of the year). That would have the consequence that every day could be slightly longer or shorter than 24h and that DST shift would occur every day.  More over the time of DST shift would be variable since sunrise is.

And if we care to be accurate to the millisecond, you need to know the actual longitude since that time variation of sunrise is a continuous phenomena and therefore the time of shift of DST in the east of the country would be different than the one in the west of the country with respect to UT for any given day.

On 12/09/2018 4:05 PM, Paul.Koning@dell.com wrote:

What does the TZ database mean "local time" to be?  Time in its common representation where zero is midnight and 12 o'clock is noon?  Or is it meant to account also for local conventions that zero is some point in the day different from midnight?

If the former, then this issue is out of scope.  If the latter, then it suggests there might be two Ethiopia zones, one for "midnight origin" (the one we have now) and one for "local convention" which combines the offsets from latitude, and the offset from the different convention of what the starting point is.

	paul

--


Patrice Scattolin | Software Development Manager | 514.905.8744
Oracle Cloud
600 Blvd de Maisonneuve West
Suite 1900
Montreal, Quebec