Talk:30W Time Zone Rule

Fresh discussion (2025)

Just summarising some discussion from the Discord.

The 30W rule is very problematic for places between the 180 antimeridian and the International Date Line, such as Tonga and Samoa. In these places, on weekdays, the geohash comes out after the midnight deadline has already expired. (Similar to the problem the 30W originally tried to solve for most of the world).

Some potential solutions and pitfalls were discussed:

1. Extend "east of 30W" to mean east all the way to the IDL. This creates situations where there are two geohashes for a given date in a given part of a split graticule, and others where there are no geohashes at all for a given date.
2. As with 1, but redraw the IDL to follow whole graticule boundaries. This is pretty awkward and arbitrary.
3. Extend the definition of "valid" to guarantee a minimum amount of time between when a geohash is published, and when it expires, local time. This could be:
   1. a number of hours, say 12 or 24, with the downside that each graticule would have a non-midnight expiry time, inconsistent with the rest of the world.
   2. a threshold at which the expiry time is pushed back to the following midnight.
   3. a blanket statement that the expiry time is the midnight of either the date the geohash is published for (nominal date) or the local day when it is published (local date)

A side effect of these options is that on weekends, a Saturday would likely have two possible points (Friday's being still valid, and Saturday having been published early enough to be achievable), while Monday would have none (its point not being available until Tuesday).

This isn't an urgent problem perhaps, but I just wanted to record the discussion. Stevage (talk) 23:28, 1 January 2025 (UTC)