May I as a layman put in just some little points: [as this is an important "initial research project" that may be used in the future by generations of Muslims]
[1] the " mushaf of Uthman" is a historical piece of evidence to indicate the authenticity of the printed "rasm Uthman" widely in use. therefore it is important to distinguish between the original copy [mushaf uthman] which DOES NOT contain diacritical marks and the printed version widely in used "rasm Uthman" which contains diacritical marks. [2] for "computational purposes" which copy is best suited to give results of the experimentations ? It is the "pure" scripts or the diacritical scripts [which was added later in the Muslim Civilisation] or it makes no difference. [3] therefore, would it be in the traditions of Islamic scholarship to encourage researchers, to "put things in their proper places" so that we do not lose whatever results obtained in our experiments to point to the "origins" . [NB. what is certain now is that the rasm Uthmani is in the calligraphy of one Scribe by the name of Uthman from Syria this is a far we have established whether he follows the "mushaf Uthman", we have yet to establish or rather the later mushaf which have diacritical marks added by later Khalifs]. --- On Mon, 2/8/10, Abdulbaqi Sharaf <a.m.shara...@leeds.ac.uk> wrote: From: Abdulbaqi Sharaf <a.m.shara...@leeds.ac.uk> Subject: RE: Advice Required - Printed Version of the Quranic Arabic Corpus for distribution To: "A Rahman Adnan" <arahma...@gmail.com> Cc: "Majdi Sawalha" <sc...@leeds.ac.uk>, "Eric Atwell" <csc...@leeds.ac.uk>, "Kais Dukes" <dukes.k...@googlemail.com>, "Quranic Arabic Corpus discussion" <comp-quran@comp.leeds.ac.uk> Date: Monday, February 8, 2010, 8:48 AM Assalamu Alaikum; Thanks Abdur Rahman for correcting me, indeed I found a number of peculiarities here http://www.islamweb.net/ver2/Archive/readart.php?id=14680 and thanks Majdi for initializing this issue and my apologies for creating this noise. what remains to discuss then is: for computational purpose should Qur'an be rendered in modern imla'e script (so to compare with contemporary body of Arabic texts?) or should it be preserved the old Uthmani script (and hence computational tools should be developed for this script?). wassalam Abdul-Baquee M. Sharaf PhD Student Language Technologies Group School of Computing University of Leeds UK ________________________________________ From: A Rahman Adnan [arahma...@gmail.com] Sent: 08 February 2010 16:23 To: Abdulbaqi Sharaf Cc: Majdi Sawalha; Eric Atwell; Kais Dukes; Quranic Arabic Corpus discussion Subject: Re: Advice Required - Printed Version of the Quranic Arabic Corpus for distribution Salam to all, As pointed out by Bro. Majdi, the Othmani script (ar-Rasm al-Uthmani) refers to the spelling system as used in the Mushaf Uthman. It is by sheer coincidence that the Syrian calligrapher's name is also Othman, but in fact ar-Rasm al-Uthmani is indeed so-called after 'Uthman ibn 'Affan, the third Caliph who commissioned a recension of the Quranic text. When you say the text is written in Rasm Uthmani it means every letter of every word is reproduced exactly as it is found in one of the copies of the Quran produced during Caliph Uthman's time. This can indeed produce some difficulty reading the unvowelled text, because the same word can be found spelled in different ways in different places. For example, the name Ibrahim is spelled throughout Surat al-Baqara "alif ba ra ha mim" but "alif ba ra ha ya mim" in other Suras. In fact a whole science has been developed by scholars ,"'ilm al-rasm", to study and explicate the peculiarities of the Uthmanic script. Wallahu a'alam. On Mon, Feb 8, 2010 at 10:47 PM, Abdulbaqi Sharaf <a.m.shara...@leeds.ac.uk> wrote: > Salam Majdi > > To the best of my knowledge the term "Mushaf Uthman" (the master copy of the > Qur'an preserved by khalifa Othman") is often confused with "ar-Rasm > al-Uthmani" or (the Othmani script) which is named after the syrian > calliographer "Othman taha" who is still alive and an employee at King Fahd > Complex at Median, the two names are just a co-incidence. The later > calliographer's script got promoted by King Fahd printing press where > millions of copies of this script are circulated, it hence became a very > popular script. > > Hence, I think the Uthmani script is a very modern script and close to the > traditional "naskh" script. The very old Qur'anic scripts are preserved in > museums and looks very different. This gives a flavor: > http://www.hqw7.com/imgs/images/file_69784.jpg > http://upload.wikimedia.org/wikipedia/commons/0/00/Koran-manuscript.jpg > > wassalam > > Abdul-Baquee M. Sharaf > PhD Student > Language Technologies Group > School of Computing > University of Leeds > UK > ________________________________________ > From: comp-quran-requ...@comp.leeds.ac.uk > [comp-quran-requ...@comp.leeds.ac.uk] On Behalf Of Majdi Sawalha > [sc...@leeds.ac.uk] > Sent: 08 February 2010 13:51 > To: Eric Atwell; Kais Dukes > Cc: Quranic Arabic Corpus discussion > Subject: RE: Advice Required - Printed Version of the Quranic Arabic Corpus > for distribution > > Hello Eric and Kais, > > iwant to highlight the issue of the Arabic writing style of the Qur'an. In >the Qur'an Corpus you use the Qthmani script, the same writing style since >the third Kalifa Othman Ibn Affan. this writing style is a bit different that >the modern standard Arabic script. but Islamic scholars to some reasons >continued to print the Qur'an using the same Othmani script. i found it easier >to read the Qur'an using modern standard Arabic script. as well as it is >easier to apply existing Arabic NLP tools to the modern standard Arabic >script. if you want to continue in using the same Othmani script, would you >please make another copy of the Qur'an using modern standard Arabic script. or >you might think of providing options (limited) similar to Tanzil project >http://tanzil.info/download > > best regards, > Majdi > ================ > MajdiSawalha, > School of Computing, > University of Leeds. > http://www.comp.leeds.ac.uk/sawalha > ________________________________________ > From: comp-quran-requ...@comp.leeds.ac.uk > [comp-quran-requ...@comp.leeds.ac.uk] On Behalf Of Eric Atwell > [csc...@leeds.ac.uk] > Sent: 06 February 2010 14:18 > To: Kais Dukes > Cc: Quranic Arabic Corpus discussion > Subject: Re: Advice Required - Printed Version of the Quranic Arabic Corpus > for distribution > > kais, > > I think this needs some careful thought ot present most of the info in > much smaller space. The question then becomes: what is the info people > really want, and what can be eliminated? part of the "information" > in the online version is layout to make it readily accessible to casual > users; arguably users of the printed form are not casual (or else they > would not get the full printed version ) and can work with aa compressed > representation - and may even prefer it if it means they can view > longer texts on one page. For example, the full explanation of > grammatical features takes up space but the same information is encoded > in more compact tags; casual visitors won't understand the tags, but a > repeat user might get used to this compact representation. > > This is generic, next step is to think about how exactly this applies to > tagged quran... I'll have a think about this, but you will probably get > good suggestions from others before me! > > eric > > > On Sat, 6 Feb 2010, Kais Dukes wrote: > >> Hello comp-quran members, >> >> I am writing to you all, to get some advice on a printed version of >> the Quranic Arabic Corpus. We have been receiving a lot of requests >> lately for something that users can download for offline use. Because >> I beleive that the accuracy of the grammar is getting quite >> reasonable, I am now considering this. In any case, we can always >> update what we produce as the grammar improves. >> >> What I had in mind, was a set of PDF files that could be downloaded. >> For example, perhaps 30 PDFs (1 per juz of the Quran, see: >> http://en.wikipedia.org/wiki/Juz'). My question is - I would be keen >> to find out from members of the mailing list if they think that this >> is a good idea? And if so, what would the best format be? I was >> thinking of starting with the word-by-word grammar (not the syntactic >> treebank). Perhaps starting with the information here: >> >> http://corpus.quran.com/wordbyword.jsp >> >> The problem is that if we displayed those images at the same >> resolution, with the same information, that comes out to about 7 >> Quranic words per printed A4 page. Given that there are 77,430 Arabic >> words in the Quran (according to our counting of whitespace) that >> would give 11,061 pages in total - or 368 pages per juz (i.e. 368 >> pages for each of the 30 PDF files). That doesn't sound very >> reasonable to me. If we shrink the images and text by 50% that would >> give about 5,530 pages in total. Do you think that perhaps we should >> just display part-of-speech tags to save space? Or perhaps Quranic >> researchers and students perfer the whole grammar written out in >> textual form? >> >> I'm open to any suggestions on how to best display this information in >> printed form for a set of PDFs. Any suggestions are more than welcome. >> Feel free to reply directly to the mailing list, just hit "reply all": >> >> comp-quran@comp.leeds.ac.uk >> >> Looking forward to hearing from you. >> >> Kind Regards, >> >> -- Kais Dukes >> >> Language Research Group >> School of Computing >> University of Leeds >> >> http://corpus.quran.com - The Quranic Arabic Corpus >> comp-quran@comp.leeds.ac.uk - Computional Quranic Arabic discussion list >> > > -- > Eric Atwell, > Senior Lecturer, Language research group, School of Computing, > Faculty of Engineering, UNIVERSITY OF LEEDS, Leeds LS2 9JT, England > TEL: 0113-3435430 FAX: 0113-3435468 WWW/email: google Eric Atwell > > -- Abdul Rahman Adnan, Tel: +603-55136162 (House) +6012-2153800 (Mobile)
