Tamil Discussion archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: glyph choices for font encoding-version 1.2

Dear Kalyan,

I saw the revised mapping.  I'm putting the finishing touches to the sorting algo
for Tamil tamil text.  While we can retain most of the glyphs, I suggest that we
drop the 'dot' character and have a glyph each for all the mey's of grantha. This
will avoid any kind of ambiguity in representing mey ezuththukkaL.  

Also, I think it'll be more complete to have the granthas listed along with the
other alphabets (for akara-uyirmey's and mey's).  It'll be more logical to treat
them as part of the alphabet set than to separate them as 'specials'.

One question I have for DTP font developers (on Windows at least). Most DTP apps.
by default do 'smart quotes'.  These smart quotes are in the 8-bit ASCII space.
We can live without them - question is : do all these common DTP/WP s/w allow the
user to overwrite smart-quote substitution ?  I know page maker does - but have
not played with any others.



---- Begin included message ----
Dear friends:
I have tried to revise the earlier proposed font encoding scheme
(version 1.1) taking into account the points raised by Leong Kok Yong,
Srinivasan, Nagu and others recently.
The revised version (labelled 1.2) is available under the URL

As in version 1.1, glyphs for all key tamil alphabets and modifiers
are grouped in the main part (slots 160-255). Debatable ones
such as tamil numerals, grantha letters and rarely used tamil alphabets
are now grouped in the rows 9 and 10 (slots 128-159).

The principal changes going to version 1.2 from 1.1 are as follows:

i) (As per the suggestions of Dr. Srinivasan of Canada)
slots 128, 129, 130, 141, 142, 143, 157, 158, 222 and 234 have been
reported to be "problems slots"  for Windows Software by Dr. Srinivasan.
So I have liberated these slots. 
I consider the issue of software problems at these slots as a temporary
one - cleared up with better handling of hardware and softwares.
So, in slots 141-144 are placed the glyphs for tamil alphabets 
nyu, nyU, ngu and ngU. These alphabets are not used
in tamil at all. So we will not loose anything even if these do not 
show up. I feel that we should have them for two reasons
i) slots code available for these to allow error-free sorting and 
interconversions of unicode based texts of other indian languages
and ii)   "tamil-learning" softwares be able to display these tamil
characters in print and on screen. I hate to have
a tamil font where we can show the entire alphabet on screen, whether
some of these are used today or not.

ii) (As per the suggestions of Leong Kok Yong ) has introduced the 
copyright sign (at #169) and the Registered mark at #174.

iii) glyphs for the unicode-type modifiers  for okara, Okara and
aukara varisai (at slots 147-149) are now deleted.

iv) in view of the criticisms for inclusion of old style characters such
as Raa, Naa, lai etc and also diacrtical markers to have transliterated
tamil text according to LC scheme, I have now removed them from the
main table. 
I would like to hear some discussions on the following possible options:
Slots 148-151 can be labelled as "user-defined" slots with 
option A containing the four old-style tamil characters (kokki for
  old style Ra, Naa and naa)  OR
 Option B carrying the four diacritical markers (horizontal bars and
above and below, using kerning).
OR we can simply go for option A or B and forget the other.

As before, please have a closer look and comment on points that are
yet to be refined.

---- End included message ----

Home | Main Index | Thread Index