Re: Quranic Proposal

[Mohammed Yousif <mhdyousif at gmx dot net>]

On خميس 10 يونيو 2004 17:43, Thomas Milo wrote:
> OK. Sounds good.
>
> After carefully reading your document, my conclusion is the following.
>
> The document enumerates problematic elements of text (characters) 
> and  elements of script (glyphs). 

 I disagree here, please Mr.Milo give use specific samples of what you
 think is so I can discuss it and either convince you or you convince me
 that I'm wrong.

> For instance, the special cases of tanween 
> fall in the domain of character encoding, 

> whereas special positioning of superscript alif
> as well as trailing alifs falls in the domain of script 
> rendering or font technology.

  I want to note that want can use 'font technology' to encode the Qur'an
  and make it look exactly like the hard copy.
  Actually this is being done all the time by different vendors and most of
  these products have been certificated by known Islamic organizations.
  I can send you more than one font that can encode the Qur'an using only
  characters from the ASCII range!
  But I do think that is is _wrong_ and should be changed, the fact that "it
  works" doesn't mean that it's good, we shouldn't delegate things to 'font
  technology' just because it can be done.

  About the superscript alef: 
    Point 1: Its use is exactly like U+06E6 and U+06E7, those two characters
               are already in the Unicode standard so if you must reject the
               new proposed character (proposed character 8.0) then I insist
               on adding to the proposal a new entry requesting the removal of
               U+06E6.
    Point 2: As in U+06E6 and U+06E7, the superscript alef can be used either
               high (and in that case it's treated as a transparent character)
               or as a regular character (and in that case it's _not_ treated
               as a transparent character at all and cannot be).
               I can give you samples if you need.
    Point 3: The existing superscript alef character has the property of being
               transparent and thus it's completely wrong and against the
               standard behavior to use it as a regular character (which is
               needed for the Qur'an as well as the transparent character).
    Point 4: Being a transparent character, the expected behavior of it is to
               be drawn on top of the previous character which is OK for some
               places in the Qur'an but which is very unacceptable for some
               more places and hence the need for two characters (again as
               in U+06E6 and U+06E7).
               I can give you samples if you need.


>
> All the characters you request are already available in the Unicode
> Standard, be it in a different form in some cases.
>
> This doesn't mean that we should stop this operation. The problem with
> Unicode and Qur'anic script is not the character repertoire, but the
> instructions for their use. Without such guide lines, Unicode cannot
> function as a standard.
>

  Then please list them one by one and tell us why you think they all exist
  in the Unicode Standard so we can remove those which are, or confirm
  those which aren't.

> A good example is the confusion regarding various sukun-like signs. They
> were requested, rejected or cross-assigned, while, all the time, they were
> already part of the standard, but nobody was aware of it because their
> shapes in the printed edition of the Unicode Standard are wrong. I am
> presently in the process of straightening out this detail - something the
> present proposal doesn't yet mention.
>

  I see no confusion at all, sukun like signs in the Qur'an are clearly
  defined in the unicode standard (using the same names that the
  hard copy of the Qur'an uses).
  They are:
    06DF (ARABIC SMALL HIGH ROUNDED ZERO)
    06E0 (ARABIC SMALL HIGH UPRIGHT ZERO)
    06E1 (ARABIC SMALL HIGH DOTLESS HEAD OF KHAH)
    06EC (ARABIC ROUNDED HIGH STOP WITH FILLED CENTRE)
  And any hard copy of the Qur'an uses exactly those names to describe
  their meaning.

> It seems to me that the sajda line needs to be taken care of by mark up
> similar to underscores in Latin script.
>

  Any suggestions?

> All in all, my suggestion is to produce a joint document that prescribes
> how to encode Qur'anic text in a uniform way using a combination of Unicode
> characters, well-designed fonts and adequate font technology as well as
> standard mark-up (i.e., XML).
>

  As I noted above, I don't like the idea of delegating everything to fonts as
  this is not the right thing to do.
  I can send you a "well designed" font that can display the Qur'an perfectly
  using only ASCII characters but this is not good at all.

  Every different character the Qur'an uses should exist on the Unicode
  Standard.
  I'm not requesting characters that are redundant and even if they look like
  that, they aren't actually and the Unicode Standard _does_ define characters
  that "look" similar (i.e. U+06E6 and U+06E7) but are really different as the
  case with the proposed characters.
  Not that this is different from "different glyphs" which should be handled
  by font technology. For instance to encode the Qur'an in "Warsh Rewaya",
  It will be needed to change almost all the glyphs but still using the same
  characters. (These glyphs look different, for example, Theh must have
  two dots above it "not three" and Teh must have only on dot "not two"
  and Beh shouldn't have any dots at all!)

  Please discuss my points and note those things that "in your opinion" are
  not reasonable so I can explain why they should be included as characters.

-- 
Mohammed Yousif
Egypt