Islamic Calendar Stuff

[Ali Abdin <ALIABDIN aucegypt edu>]

been browsing online for the islamic calendar stuff (and looking at the 
emacs cal-islam.elc).

First of all - if anyone has this book or can find it and bring it to 
GUADEC (e.g. from a library) for a couple of days I would appreciate it 
:) Its not at my Uni's library

The title is:
A Modern Guide to Astronomical Calculations of Islamic Calendar, Times & 
Qibla. Berita Publishing, 1984, (ISBN: 967-969-009-1)

Okay - ive also browsed online for a bit. Here is a very good site if you 
want to familiarize yourself with it:
http://www.tondering.dk/claus/cal/node5.html

This site also explains the way emacs calculates its dates. They call it 
a "crude system" and i don't fully understand it, and thats just not how 
the calendar works. 

the whole point of the islamic calendar is that its not based on the 
actual 'new moon' stuff (that would be easy to calculate) but it is based 
on the /sighting/ of the new moon. The sighting is the unpredictable part 
(and they only need the sighting during 3 times in the year, two eids and 
1st of ramadan). There is a program out there (DOS-only - no source code) 
that measures the probability of a moon sighting during a specific time 
(based on some astro code) which could be used as a basis for the 
calendar - can libastro do this? any intended functionality for it? The 
URL is: http://www.ummah.org.uk/ildl/

I'll think i'll just do the following: sync the islamic date with the 
gregorian date using a newspaper from sometime soon :) of course - this 
will cause inaccuracies when you view 'past' dates (they could be off by 
a few days). There is no way to accurately detect this unless we store 
the number of days for each month for all the prior years (well actually 
for two months only)).

Any comments?