Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!mailrus!accuvax.nwu.edu!naim
From: naim@eecs.nwu.edu (Naim Abdullah)
Newsgroups: soc.religion.islam
Subject: Re: Proposal for a Machine readable Qur'an
Message-ID: <1890@accuvax.nwu.edu>
Date: 7 Dec 89 06:01:48 GMT
Sender: news@accuvax.nwu.edu
Lines: 48
Approved: naim@eecs.nwu.edu (Naim Abdullah)

Camm Maguire proposes a project for typing in an English translation
of the Quran and asks:

>	1) Is there interest in this project?  Who would be willing
>to contribute typing time?  How much time could they afford to
>spend? 

Certainly, there would be lots of interest in this project.
You can count me in (after Jan 21st, '90), on typing 100 verses.

>	2) What English translation of the Qur'an should we choose?
>Many of you may know of the world scripture translation project
>headed by Prince Charles.  I hear that in the near future, a new
>English translation fully approved of by all leading Muslim scholars
>will be produced.  We may want to wait for this translation to be
>completed, or we may not.

How about choosing an English translation that is in modern English
rather than the ones that usually use archaic English ? In a previous
posting, Asim Jalis mentioned a translation by Ahmed Ali that uses
contemporary English.

We probably do not want to wait for the translation from the world 
scripture project. Who knows when they will finish, and whether their
result will be any better than existing translations.

>	3) We need to decide on a procedure for quality control,
>such as assigning a typer and a different proof-reader for each
>section.

There are two possibilities:

1) Somebody with access to a scanner can scan in a whole translation.
This would ensure accuracy and also capture the Arabic text. However
the resultant output would be a bitmap image rather than ASCII text.
It would only be viewable on workstation monitors.

2) If we stick with the approach of an army of typists, then if we collect
126 (= 63 * 2) people, we can have two people independently do every
section. Every pair of outputs can be compared using UNIX text processing
tools to detect differences (there are variants of diff around that can do
this). If we can barely collect 63 people, then we can proofread each others
work.

What do other people think of Camm's proposal ? Here is a chance to volunteer
and make a difference!!

     Naim