Generally it is within a day. Off the top of my head I cannot say whether the error always lies on one side, but the calcs could be tweaked to always be on the safe side, however that may be defined. Tim Smartcom Software Ltd Portsmouth Technopole Kingston Crescent Portsmouth PO2 8FA United Kingdom www.smartcomsoftware.com Smartcom Software is a limited company registered in England and Wales, registered number 05641521. -----Original Message----- From: tz-bounces@iana.org [mailto:tz-bounces@iana.org] On Behalf Of Ian Abbott Sent: 16 May 2012 11:29 To: tz@iana.org Subject: Re: [tz] Calculating dates of Ramadan On 2012-05-16 10:28, Tim Thornton wrote:
Regarding calculating Ramadan: The Islamic calendar is lunar, not solar, and is about 11 or 12 days shorter than the "normal" Gregorian calendar. Months (including Ramadan) start when the first crescent of a new moon is sighted. This varies according to the atmospheric conditions and the exact positions of earth, sun and moon, but can be calculated to within about a day for any given location, which is good enough for tz data in the absence of any further specific information for a location. There is also an ongoing debate in Islam on whether the calculated date is good enough, or whether they should stick with past practice of having official observers watching the sky to announce the start of the month. What cannot be done is to say that it always falls between a given pair of Gregorian calendar dates, due to its shifting.
Tim
In practice, how much does the observed (as in astronomical) date differ from the calculated date? Does it ever start _before_ the calculated date? -- -=( Ian Abbott @ MEV Ltd. E-mail: <abbotti@mev.co.uk> )=- -=( Tel: +44 (0)161 477 1898 FAX: +44 (0)161 718 3587 )=-