Thanks for letting us know. This means TZDB will need to change for the spring-forward transitions in 2024 and 2025. We might as well adjust predictions for later years to match, though of course this is just guesswork. I installed the attached proposed patch to TZDB to do all that. We hope to have a new TZDB release soon. Given the short notice, likely many cell phones and computers will not be updated in time. It might help if you spread the word that people whose devices are not yet updated can work around the problem by manually setting their devices' time zones to Cairo time on April 13, and manually setting them back to Palestine time on April 20. This sort of hassle can be avoided in the future by giving longer notice of rule changes.

On 2024-02-03 13:06, brian.inglis--- via tz wrote:

There should be some better sources available of accurate astronomical moon sighting calculations

The timestamps TZDB documents are civil timestamps, not religious ones. This is why TZDB cites the Palestinian Cabinet's predictions, not the astronomical observations that are the basis of the Islamic calendar. Although I don't know, I guess that start-of-Ramadan crescent moon sightings in Palestine are evaluated by an investigation committee of the Palestinian Fatwa House <https://www.darifta.ps/>. If so, surely the Cabinet would defer to the Fatwa House if the Cabinet's conservative predictions of a period of time around Ramadan turn out to be not conservative enough. However, if this (dare I say it?) astronomically improbable event occurs, that won't change the start and end of Ramadan - all it would mean is that legal clock changes wouldn't match Ramadan that year, due to the Cabinet's mistake. At most the Cabinet would make some last-second correction (perhaps even an ex-post-facto correction!) and tzdata could continue to cite just the Cabinet. This is because TZDB's goal is to document what human clocks say, not what the earth and moon are doing.