indic-computing-devel — Discussing the development of tools for Indian language information processing

[Indic-computing-devel] kannada keymap

[<pra...@ya...>]

Greetings group,
I am currently tinkering with translating KDE into
kannada. I translated a few kde POT files (into
kannada) using yudit & could view it with kbabel. 
A few questions:
I used Code2000 & since it is a shareware, i am
looking for another font. Any suggestions?
The default kannada keymaps that came with yudit are
sorely inadequate & don't display all the Consonants
like. Try typing "kannada" & you will see. 
Can i convert the yudit keyboard layout sources to any
other format, such that i can use it with other apps,
like kbabel?  
Thanks in advance,
pramod.



________________________________________________________________________
For live cricket scores download  Yahoo! Score Tracker
 at: http://in.sports.yahoo.com/cricket/tracker.html

Re: [Indic-computing-devel] CDAC swallows NCST

[Keyur S. <key...@ya...>]

--- "Tapan S. Parikh" <ta...@ya...> wrote:

> There is a move to
> merge NCST with CDAC. If this happens, all the good works

> done towards public
> domain free software initiatives of NCST (Indian Language
> Linux, OpenOffice, etc)will come to a standstill. This is

> quite obvious as CDAC wants to make money by selling 
> Indian language fonts, keyboard
> drivers, office suits, etc (they are funded by Govt, is
> another matter).

I have recently back to Mumbai after my Delhi visit. The
visit was for ELITEX-2002 (http://www.elitexindia.com)
exhibition. There I met many dignitaries. I have just made
it sure that all work done by NCST including Linux and
OpenOffice localization will remain in public domain. So
guys, don't worry about it. And as far as I know, nothing
has finalized about merging of CDAC with NCST.

> > Someone should educate [snip]

Please do not write or encourage anything offensive to
someone.

Regards,
Keyur



__________________________________________________
Do You Yahoo!?
Yahoo! Health - your guide to health and wellness
http://health.yahoo.com

[Indic-computing-devel] Fw: [STSF-announce] Standard Type Services Framework (STSF) 0.2 Released

[Guntupalli K. <kar...@fr...>]

Begin forwarded message:

Date: Tue, 30 Apr 2002 19:20:26 -0700
From: Alexander Gelfenbain <ad...@su...>
To: sts...@li...
Subject: [STSF-announce] Standard Type Services Framework (STSF) 0.2
Released



The STSF team announces the 0.2 release of the Standard Type Services
Framework.

The source code and FreeBSD binary packages (Linux RPM packages to
follow) are available for download from
http://sourceforge.net/projects/stsf/

STSF consists of two parts:

ST  - rendering-system independent framework

XST - X11 extension to XFree86 4.2.0 that exports ST functionality
          to X11 clients

ST compiles on Solaris, Linux, and FreeBSD platforms.  XST compiles on
Solaris, Linux, and FreeBSD as well, but does not contain drivers for
common Solaris/sparc graphics cards, so is not usable on most
Solaris/sparc platforms at this time.

The source code comes in two archives:
        st-0.2.src.tgz
                contains the entire ST source tree

        xst-0.2_xfree86-4.2.0.src.tgz
                contains the entire XFree86 4.2.0 + XST extension
                source tree

        Alternatively, if you already have XFree86 4.2.0 source
        tree you can apply a patch xfree86-xst-0.2.patch.gz to
        it.

ST currently requires GNU make and FreeType
(See the docs/INSTALL file in the ST distribution for full
instructions.)

XST requires ST, Motif or OpenMotif, and GNU libiconv if your system
libc does not provide iconv.  (A link to the ST source tree must exist
as "ST" in the parent directory of the XST tree.)

Neither ST nor XST are completely implemented at this time, but you
can see what is available by running our xclock -digital or xstfontsel
demos in X or the ST web demo cgi using the GD graphics library.

Mailing lists for comments, questions, or further discussion are
available at:        http://sourceforge.net/mail/?group_id=46276

The STSF team.


-- 
Alexander Gelfenbain, Sun Microsystems, Inc.
+1 (408) 635-0612 

[Indic-computing-devel] Fw: [bytesforall_readers] CDAC swallows NCST

[Tapan S. P. <ta...@ya...>]

----- Original Message -----
From: "Kannada Ganaka Parishat" <ka...@em...>
To: <byt...@ya...>
Sent: Thursday, May 02, 2002 10:22 AM
Subject: [bytesforall_readers] CDAC swallows NCST


> Hi all,
>
> After the voluntary retirement of Dr. S. P. Mudur as Director of
NCST (National Centre for Software Technology), no new director has
been appointed. Dr Arora, Director of CDAC has been given the
additional charge as Director of NCST. There is a move to merge NCST
with CDAC. If this happens, all the good works done towards public
domain free software initiatives of NCST (Indian Language Linux,
OpenOffice, etc)will come to a standstill. This is quite obvious as
CDAC wants to make money by selling Indian language fonts, keyboard
drivers, office suits, etc (they are funded by Govt, is another
matter).
>
> Someone should educate Mr Pramod Mahajan, Hon'ble minister for ICT,
GoI.
>
> -SS
> --
>
> _______________________________________________
> Sign-up for your own FREE Personalized E-mail at Email.com
> http://www.email.com/?sr=signup
>
>
>
> ------------------------ Yahoo! Groups
Sponsor ---------------------~-->
> Tied to your PC? Cut Loose and
> Stay connected with Yahoo! Mobile
> http://us.click.yahoo.com/QBCcSD/o1CEAA/sXBHAA/C7EolB/TM
> --------------------------------------------------------------------
-~->
>
> To unsubscribe from this group, send an email to:
> byt...@ya...
>
>
>
> Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
>

Re: [Indic-computing-devel] Telugu OTL font

[Arun S. <ar...@sh...>]

Keyur Shroff wrote:

>
>I got response from Vijay, Murali, Karunakar, Nagarjun. The
>font is available at
>
>http://rohini.ncst.ernet.in/indix/download/font/
>
>I got the font from Ganga Prasad <gan...@em...>,
>who is ready to put the font in the public domain under
>GPL.
>

"Public domain" and GPL are mutually exclusive, but both fall under the 
category of free software (in this case, free fonts).

http://www.gnu.org/philosophy/categories.html

    -Arun

[Indic-computing-devel] (Fwd) [LRegroup][115] Admission to Research Programmes in Comp

[Dr. U.B. P. <pav...@vi...>]

------- Forwarded message follows -------
Date sent:      	Wed, 24 Apr 2002 19:23:34 +0530 (IST)
From:           	Lexical Resource Egroup <lr_...@ii...>
To:             	Lexical Resource Egroup <lr_...@ma...>
Subject:        	[LRegroup][115] Admission to Research Programmes in Computational
	Linguistics at Intl.IIT, Hyderabad. 

[LRegroup][115] Admission to Research Programmes in Computational
Linguistics at Intl.IIT, Hyderabad

                 LANGUAGE TECHNOLOGIES RESEARCH CENTRE
      International Institute of Information Technology, Hyderabad 
                        (Deemed University)
          [formerly Indian Institute of Information Technology]       
                 


    ANNOUNCEMENT OF RESEARCH PROGRAMMES IN COMPUTATIONAL LINGUISTICS


    Language Technologies Research Centre (LTRC) announces two post
    graduate 
 research programmes in Computational Linguistics

   - 1 Year Post Graduate Diploma in Computational Linguistics

   - PhD (Computational Linguistics)


   The programmes primarily involves research in both theoretical and
   applied 
  aspects of natural language processing studies. Research areas being
  pursued involve morphology and grammar frameworks and their parsers,
  machine learning of grammar and lexical properties from corpora,
  preparation of  lexical resources, development of architecture for
  information extraction and retrieval, text classification, speech
  processing and relating Western  language theories with Paninian
  grammar and Navya-nyaya. The major thrust   of research is to build
  theoretical models which can be used for computer applications and
  investigate how language knowledge can be engineered to develop
  tools in areas such as machine translation, information extraction
  and speech synthesis. 


 1. ADMISSION TO RESEARCH PROGRAMMES

 1.1 Eligibility

   * 1 Year  P.G. Diploma in  Computational   Linguistics :
   * PhD(Computational Linguistics): 


     Both programmes consist of two streams.

    Stream 1: Candidates with Mathematics, Science or Engineering
              background (such as B.Tech/B.E. in any discipline,
              M.A./M.Sc. in Mathematics or Statistics, M.Sc. in
              Physics) and having a keen interest or aptitude for
              language analysis and processing can apply. The course
              is meant for only those students who have aptitude for
              language analysis.

    Stream 2: Candidates with background in Linguistics or languages,
              Sanskrit students or individuals with an aptitude in
              language analysis are encouraged to apply.


 1.2 Financial support for Research Programmes

    A number of research assistantships covering full tuition and
    living 
   expenses are available for PhD programme. 


 2. STRUCTURE OF RESEARCH PROGRAMMES 

 2.1. 1 Year Post-Graduate Diploma in Computational Linguistics

    Students are required to do 6 courses and a project. The courses
    are 
  selected  from the list of courses offered by LTRC. Subject to
  satisfactory performance, students can continue for PhD
  (Computational Linguistics).

 2.2. PhD (Computational Linguistics)

    Students are required to do the following minimum courses (7 in
    number) 

   * Programming - 1 (Scripting) 
   * Programming - 2 (Data Structure, DBMS) 
   * Probability & Statistics 
   * Logic 
   * Natural Language Processing 1 
   * Natural Language Processing 2 
   * Linguistics (half semester course) 
   * Indian Grammatical Tradition (half semester course) 

   They must pass a comprehensive examination in Computational 
   Linguistics, and produce a research thesis which advances the
   state of the art.

   A number of courses listed above are specially designed to cater to
   specific needs of students with background in linguistics,
   Sanskrit, and languages. 

 3. APPLICATION

    The application form can be downloaded from 
    http://www.iit.net/PG_CompLing.htm

    LAST DATE FOR SUBMISSION OF APPLICATION FORM : MAY 15, 2002


    APPLY DIRECTLY TO:

    Language Technologies Research Centre
    International Institute of Information Technology
    (Deemed University)
    [formerly Indian Institute of Information Technology] 
    Gachibowli, Hyderabad 500 019, India

    e-mail: lt...@ii...
    Phone : +91-40-300 1412
    +91-40-300 1967 Extension 144
    FAX : +91-40-300 1413
    URL : http://www.iiit.net/ltrc/index.html

======================================================================
======== Electronic Group: Lexical Resources for Natural Language
Procesing

Web: http://www.iiit.net/lr/index.html
Email addr for LR egroup: lr_...@ii...





------- End of forwarded message -------
-----------------------------------------------------
Dr. U.B. Pavanaja
Editor, Vishva Kannada
World's first Internet magazine in Kannada
http://www.vishvakannada.com/

Note: I don't worry about pselling mixtakes

[Indic-computing-devel] Re: drutvamu re-ordering

[Ganga P. <gan...@em...>]

From: Keyur Shroff <key...@ya...>
> actually follows Ka. So I think actual reordering should be
> 
>    Sa Ka Halant VowelSignVocalicR

exactly

 
> I still don't understand why you are reordering
> VowelSignVocalicR before Ka and not after it. After
> reordering, position of the glyphs matches 

not me - uniscribe engine. But Dr.Desikachari (pothana2000 font designer) and Microsoft (Apoorva Joshi's team) some how - with some rule they are swapping their places. Anyway, I did communicated with Dr.Desikachari, he promised me to look into it and get back to me.

...
> 
> And one more erquest to you. Why don't you join
> indic-computing-devel list? You can visit homepage at
> http://indic-computing.sourceforge.net/
> and you can become member by visiting following page :
> http://sourceforge.net/mail/?group_id=38057
> 
> As a member of the team, you can better communicate with
> the whole group.

I just joined.

Regards,
GangaPrasad

-- 

_______________________________________________
Sign-up for your own FREE Personalized E-mail at Email.com
http://www.email.com/?sr=signup

[Indic-computing-devel] Re: drutvamu re-ordering

[Keyur S. <key...@ya...>]

Hi,

--- Ganga Prasad <gan...@em...> wrote:

> > Input sequence :
> > Sa Halant Ka Halant Ra VowelSignU
> 
> I disagree with the above input sequence - it supposed to
> be (for Telugu it is bit different from Hindi, in telugu
> always first sound of the syllable will take full form
> and followed by 2nd/3rd consonant sound(s) form as half
> form(s) - in hindi it is not quite - depends on sounds;
> for word 'prasad' in both hindi and Telugu pa will have
> full form, where as in word 'kaniSSka', syllable SSka -
> in telugu - as usual SSa is full and k is half form, but
> in hindi it is otherway around)

Oh! Now I have started learning Telugu also. ;-)

> 
> Sa Halat Ka VOWEL SIGN VOCALIC R  (there won't be halat
> for Ka and no Ra)

Yes, the same sequence we use for Hindi also. But as U
mentioned Vattu, I thought that U mis-spelled it :-)

> and re-ordering for telugu as below
> 
> Sa VOWEL SIGN VOCALIC R Ka Halat 
> and then rules go.

After discussing these issues with one of my colleague here
who knows telugu, I have come to know that
VowelSignVocalicR (which is an exception in vowel signs)
actually follows Ka. So I think actual reordering should be

   Sa Ka Halant VowelSignVocalicR

I still don't understand why you are reordering
VowelSignVocalicR before Ka and not after it. After
reordering, position of the glyphs matches with their
visual order. As you said, VowelSignVocalicR should be
displayed after Ka and hence it should also be reordered
like that. You should swap position of all matras except
VocalicR with half form.

> Here problem is with re-ordering of VOWEL SIGN VOCALIC R
> with any half forms. Other vowelsigns, like VowelSignU
> (which is also positioned on right side of the consonant
> but attached to it. Where as VOWEL SIGN VOCALIC Rs placed
> on right side, but not attached to consonants; rest of
> the vowelSigns are placed on top of the consonant as
> attaced, with exception to vowelSign_AI which is not
> attached but will be placed on top and bottom - two
> glyphs) doesn't have problem since they are dispayed as
> they are after Uniscribe re-ordering. But with VOWEL SIGN
> VOCALIC R, we need to swap half forms position with VOWEL
> SIGN VOCALIC R; as a result the syllable will be (and
> should be) as
> 
> Sa half_form_Ka VOWEL SIGN VOCALIC R

This is what my friend here at NCST has explained me. Now I
can understand whatever you have explained above.


And one more erquest to you. Why don't you join
indic-computing-devel list? You can visit homepage at
http://indic-computing.sourceforge.net/
and you can become member by visiting following page :
http://sourceforge.net/mail/?group_id=38057

As a member of the team, you can better communicate with
the whole group.


Regards,
Keyur




__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

[Indic-computing-devel] drutvamu re-ordering

[Ganga P. <gan...@em...>]

----- Original Message -----
From: Keyur Shroff <key...@ya...>
> hi,

Furst - I'm sorry that I confused you using word vattu instead of halat (that's my mistake, in telugu we call it as vattu for halat)


> Word : Samskruthi
> Syllable : Skru
> 
> Input sequence :
> Sa Halant Ka Halant Ra VowelSignU

I disagree with the above input sequence - it supposed to be (for Telugu it is bit different from Hindi, in telugu always first sound of the syllable will take full form and followed by 2nd/3rd consonant sound(s) form as half form(s) - in hindi it is not quite - depends on sounds; for word 'prasad' in both hindi and Telugu pa will have full form, where as in word 'kaniSSka', syllable SSka - in telugu - as usual SSa is full and k is half form, but in hindi it is otherway around)

Sa Halat Ka VOWEL SIGN VOCALIC R  (there won't be halat for Ka and no Ra)

> 
> Reordered sequence :
> Sa Halant Ka Ra Halant VowelSignU

and re-ordering for telugu as below

Sa VOWEL SIGN VOCALIC R Ka Halat 

and then rules go.

Here problem is with re-ordering of VOWEL SIGN VOCALIC R with any half forms. Other vowelsigns, like VowelSignU (which is also positioned on right side of the consonant but attached to it. Where as VOWEL SIGN VOCALIC Rs placed on right side, but not attached to consonants; rest of the vowelSigns are placed on top of the consonant as attaced, with exception to vowelSign_AI which is not attached but will be placed on top and bottom - two glyphs) doesn't have problem since they are dispayed as they are after Uniscribe re-ordering. But with VOWEL SIGN VOCALIC R, we need to swap half forms position with VOWEL SIGN VOCALIC R; as a result the syllable will be (and should be) as

Sa half_form_Ka VOWEL SIGN VOCALIC R

(and positing of half_form_Ka will again depends on stle, since in telugu there are 2 different styles, I'm using traditional old style in bharathi font - which should be placed adjusant to full form Consonant, but when I write letters to my parents/siblings/friends I use 2nd style - which is plced below the full form Consonant)


> 
> Step-1 : Apply 'half' feature
> HalfSa Ka Ra Halant VowelSignU
> 
> Step-2 : Apply 'blwf' feature
> HalfSa Ka Vattu VowelSignU
> 
> Step-3 : Apply 'vatu' feature
> HalfSa Kra VowelSignU
> 
> Step-4 : Apply GPOS features ...


Though Hindi and Telugu belongs to Indic family and they almost have same number of character set, but their spelling is quite different. Most of the time the rules used for hindi (with reference to OTL) will not work for Telugu (and same to kannada too)

Guys, it is getting interesting and getting better and better, I spent more than 6 months alone on this one and 2 months ago I scrapped all my old work and re wrote the rules. Now with these discussions, I feel we will be reaching our goal soon and with better quality. Hope more people will follow the lead and publish more fonts in various languages.

Thank you for your time and efforts.
GangaPrasad

-- 

_______________________________________________
Sign-up for your own FREE Personalized E-mail at Email.com
http://www.email.com/?sr=signup

[Indic-computing-devel] Re: About your Telugu font (Bharathi)

[Keyur S. <key...@ya...>]

hi,

--- K Nagarajan <kn...@wi...> wrote:

> 
> > About Bha - you are  correct,  till now I put all my
> efforts  on fixing the
> > rules, so that syllables are rendered correctly (except
> for drutvamu - like
> > in  word -  samskruthi,  where  in 2nd  syllable  it 
> supposed  to  display
> > sa+ka_vattu+vowelSignRU,      but     this      font   
> is      displaying
> > sa+VowelSignRU+ka_vattu;  in VOLT proof reader it
> displays correctly but in
> > real app, such as wordpad, IE it is not)

This is probably because you have encoded rules without
considering reordering of character. Let me give an example
with Devanagari script :

When we type Ka Halant Ra, then it turns into 'Kra', with
Ka as base glyph and Vattu glyph attached to it. However,
the rules for vattu is put as in GSUB table :
'Ra Halant -> Vattu' instead of 'Halant Ra -> Vattu'

This is required to minimize the number of all
theoratically possible combination rules. So in VOLT proof
reader, you have to type the sequence after reordering,
viz., 'Ka Ra Halant' to check rules for the syllable 'Kra'.

Also the word Sanskruthi in Hindi/Sanskrit is written using
VOWEL SIGN VOCALIC R attached to KA. Vattu is not used.
However if you use Vattu then it should be formed by the
following substitution rules :

Word : Samskruthi
Syllable : Skru

Input sequence :
Sa Halant Ka Halant Ra VowelSignU

Reordered sequence :
Sa Halant Ka Ra Halant VowelSignU

Step-1 : Apply 'half' feature
HalfSa Ka Ra Halant VowelSignU

Step-2 : Apply 'blwf' feature
HalfSa Ka Vattu VowelSignU

Step-3 : Apply 'vatu' feature
HalfSa Kra VowelSignU

Step-4 : Apply GPOS features ...

This way, you get the glyhs for HalfSa, Kra, and
VowelSignU.

There are alternative ways to substitue glyphs, but the
abovementioned is cleaner approach used in quality fonts.

> I guess we can have a somewhat bigger font (more memory)
> rather than lesser quality.

I agree. I also prefer medium-range font having number of
glyphs between 200 and 500.

> Thanks. I will see if I can compact the font with just
> Telugu in it, otherwise I will ask your help :-)

Be careful! If you add or delete any glyph in the font then
all the opentype rules will be disturbed and font will
become useless. So you should just remove mapping of
Unicode values to the glyphs that you don't want. In
OpenType font, tables should be added to the font ONLY
after you finish with the glyph set; otherwise it can
create many problems for you. Take backup of your font
before doing any experiment.

Regards,
Keyur



__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

[Indic-computing-devel] Re: About your Telugu font (Bharathi)

[Keyur S. <key...@ya...>]

Note: forwarded message attached.


__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

[Indic-computing-devel] Re: About your Telugu font (Bharathi)

[Keyur S. <key...@ya...>]

Note: forwarded message attached.


__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

[Indic-computing-devel] About your Telugu font (Bharathi)

[Keyur S. <key...@ya...>]

Note: forwarded message attached.


__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

[Indic-computing-devel] Sui and Damaru!!!

[Rajkumar S <s_...@my...>]

Hi all,

This is an interesting extract from
http://www.greenstar.org/butterflies/Hole-in-the-Wall.htm, where the
author talks about an experiment involving introducing interent to slum
childern.

Definitly worth reading.

raj

Q: But most of the information on the Internet is in English and the
people you're talking about don't speak English.

A: We had some very surprising results there. We all have great
misconceptions about what these children know and don't know. At first, I
made a Hindi interface for the kids, which gave them links for hooking up
with Web sites in their own language. I thought it would be a great hit.
Guess what they did with it? They shut it down and went back to Internet
Explorer. I realized that they may not understand the dictionary meaning
of [English] words, but they have an operational understanding. They know
what that word does. They don't know how to pronounce F-I-L-E, but they
know that within it are options of saving and opening up files ...

The fact that the Internet is in English will not stop them from accessing
it.

They invent their own terminology for what's going on. For example, they
call the pointer of the mouse sui, which is Hindi for needle. More
interesting is the hourglass that appears when something is happening.
Most Indians have never heard of an hourglass. I asked them, "What does
that mean?" They said, "It's a damru," which is Hindi for Shiva's drum.
[The God] Shiva holds an hourglass - shaped drum in his hand that you can
shake from side to side. So they said the sui became a damru when the
"thing" [the computer] was doing something.

[Indic-computing-devel] Test. Please Ignore

[Harsha R. <har...@ya...>]

and delete.

-Harsha


__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

Re: [Indic-computing-devel] Telugu OTL font

[Keyur S. <key...@ya...>]

--- "Tapan S. Parikh" <ta...@ya...> wrote:
> 
> No, but a good friend of mine does.  CC'ed here.  Vijay,
> Murali Medisetti -
> can you help Keyur with this?
> 
> I think Karunakar may also be Telugu, but he went back to
> Hyderabad
> today....  Must be other Hyd based people around...

I got response from Vijay, Murali, Karunakar, Nagarjun. The
font is available at

http://rohini.ncst.ernet.in/indix/download/font/

I got the font from Ganga Prasad <gan...@em...>,
who is ready to put the font in the public domain under
GPL.
The name of the font is "bharathi" and he has requested to
keep the name while we distribute it under GPL. So we'll
keep the same name and make sure that he'll get credit for
his work.

Please test the font and give your feedback to either me or
directly to Ganga Prasad. We can also put this font in the
repository of indic-computing website from where people can
download it. If anyone of you is ready to do further work
voluntarily on this font then it will be great.

Regards,
Keyur



__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

Re: [Indic-computing-devel] Telugu OTL font

[Vijay P. S. A. <vi...@sr...>]

Send the font here (my mail id)  keyor, Murali can see it...

vijay

----- Original Message -----
From: "Tapan S. Parikh" <ta...@ya...>
To: "Keyur Shroff" <key...@ya...>;
<ind...@li...>
Cc: <med...@ya...>; "Vijay-Sristi" <vi...@sr...>
Sent: Thursday, April 11, 2002 9:01 PM
Subject: Re: [Indic-computing-devel] Telugu OTL font


>
> No, but a good friend of mine does.  CC'ed here.  Vijay, Murali
Medisetti -
> can you help Keyur with this?
>
> I think Karunakar may also be Telugu, but he went back to Hyderabad
> today....  Must be other Hyd based people around...
>
> --Tapan
>
> ----- Original Message -----
> From: "Keyur Shroff" <key...@ya...>
> To: <ind...@li...>
> Sent: Thursday, April 11, 2002 8:40 PM
> Subject: [Indic-computing-devel] Telugu OTL font
>
>
> > Hi,
> >
> > Is there anyone on this list who knows Telugu? Someone has
> > given me an opentype Telugu font. It will be great if we
> > can test the font and put it in public domain for free
> > download.
> >
> > - Keyur
> >
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Yahoo! Tax Center - online filing with TurboTax
> > http://taxes.yahoo.com/
> >
> > _______________________________________________
> > Indic-computing-devel mailing list
> > http://indic-computing.sourceforge.net/
> > Ind...@li...
> > https://lists.sourceforge.net/lists/listinfo/indic-computing-devel
>

Re: [Indic-computing-devel] Telugu OTL font

[Tapan S. P. <ta...@ya...>]

No, but a good friend of mine does.  CC'ed here.  Vijay, Murali Medisetti -
can you help Keyur with this?

I think Karunakar may also be Telugu, but he went back to Hyderabad
today....  Must be other Hyd based people around...

--Tapan

----- Original Message -----
From: "Keyur Shroff" <key...@ya...>
To: <ind...@li...>
Sent: Thursday, April 11, 2002 8:40 PM
Subject: [Indic-computing-devel] Telugu OTL font


> Hi,
>
> Is there anyone on this list who knows Telugu? Someone has
> given me an opentype Telugu font. It will be great if we
> can test the font and put it in public domain for free
> download.
>
> - Keyur
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! Tax Center - online filing with TurboTax
> http://taxes.yahoo.com/
>
> _______________________________________________
> Indic-computing-devel mailing list
> http://indic-computing.sourceforge.net/
> Ind...@li...
> https://lists.sourceforge.net/lists/listinfo/indic-computing-devel

[Indic-computing-devel] Telugu OTL font

[Keyur S. <key...@ya...>]

Hi,

Is there anyone on this list who knows Telugu? Someone has
given me an opentype Telugu font. It will be great if we
can test the font and put it in public domain for free
download.

- Keyur



__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

[Indic-computing-devel] Microsoft Localisation

[Tapan S. P. <ta...@ya...>]

Just a quick note to tell you guys that Venky and I met the head of
localisation efforts in India for Microsoft yesterday.  Great guy, very
open to collaboration and discussion.  Ive invited him and his technical
staff to join our lists.  One really great thing I think they will be able
to contribute is that they have been building a Hindi lexicon for interface
terms, in collaboration w/ several local companies in India.  If they share
this with us it would be a big help for app developers to use common terms
in our interfaces w/ those in MS, which as we all know, will be very
important in the long term.

Im sure other good technical interactions will arise also, as Im sure we
are all interested in how MS is approaching localisation.  Lets hope they
join and get active.

--Tapan

Re: [Indic-computing-devel] Yudit 2.5.4

[Rajkumar S <s_...@my...>]

On Fri, 29 Mar 2002, Dr. U.B. Pavanaja wrote:

> If download it for Windows, can I get Unicode Kannada editor on Win98?
> What about the problem of Win98 NOT supporting OpenType fonts?  Has it
> got a rendering engine for Indic scripts for Opentype fonts?

Yudit do not depend on the native OS for font support. I have't used it
under any windows, but in linux it works beautifully.

> I did actually download it for Windows and tried both on Win98 and
> WinXP. I selected Kannada. Entered some unicode values for Kannada.
> But I did not see Kannada text. On XP I do have Kannada Opentype
> fonts.

You need to select an open type font which can support Kannada I think
code2000 or ArialUni has kannada support, then you need to select an input
method for kannada, I think an inscript input method is present in yudit.
Once this two steps are completed you should be able to see kannada in
yudit.

Please get back to me if any problems happens.

raj

Re: Letter frequency in unicode (Was Re: [Indic-computing-devel] Free UCS outline font)

[Dr. U.B. P. <pav...@vi...>]

> People at Prajavani (a leading Kannada daily) have done the frequency
> > analysis for Kannada letters looooooooong ago. They had even
> designed > their Montype Kannada layout based on these data.
> 
> Did they publish their findings somewhere ?

I tried to hunt for it. But could not get hold. So many people have 
changed during these many years at their technical section and nobody 
knows about the whereabouts of that data.

> 
> > Now all these have 
> > gone due to the advent of GoK (Govt of Karnataka) standard Kannada
> > keyboard layout (also know as KGP (Kannada Ganaka Parishat) std).
> 
> Did the KGP keyboard take into account such analyses ? I'm sure it's
> easy to use for people already comfortable with qwerty.

No. KGP/GoK keyboard is based on QWERTY. Except for Z,X,Q,W and the f 
(link key) keys all othe rkeys are intuitive for those who are 
familiar with QWERTY. For ex, k is for ka, K is for Kha, etc. For 
those people who use Englsih as well as Kannada, this keyboard layout 
becomes very easy.

KGP layout is not totally invented by KGP. It was originally by KP 
Rao and was modified slightly by KGP. KP Rao is 80+ years of age now, 
but still active. He is currently with his daughter in Delhi. I am in 
touch with him about our common interests.

> If the world didn't have qwerty or English and you were (the Indian)
> Sholes designing a keyboard for kannada, how would you design it ?

Probably I would have designed it very close to the Monotype layout 
once used by Prajavani. 

> > I hope everyone remembers the story about QWERTY keyboard that we
> > use for English. It was designed to make typing SLOW rather than
> > FAST ;-)
> 
> Qwerty isn't the unchallenged king of the keyboard universe. I
> recently bumped into Dvorak:

I am aware of another layout by IBM. It was called something EBC... I 
had used it during my card-punching days at BARC. It was far superior 
to qwerty, but it died.

Rgds,
Pavanaja
-----------------------------------------------------
Dr. U.B. Pavanaja
Editor, Vishva Kannada
World's first Internet magazine in Kannada
http://www.vishvakannada.com/

Note: I don't worry about pselling mixtakes

Re: [Indic-computing-devel] Yudit 2.5.4

[Arun S. <ar...@sh...>]

On Fri, Mar 29, 2002 at 01:31:22PM +0530, Dr. U.B. Pavanaja wrote:
> Hi,
> 
> If download it for Windows, can I get Unicode Kannada editor on 
> Win98? What about the problem of Win98 NOT supporting OpenType fonts? 
> Has it got a rendering engine for Indic scripts for Opentype fonts?
> 
> I did actually download it for Windows and tried both on Win98 and 
> WinXP. I selected Kannada. Entered some unicode values for Kannada. 
> But I did not see Kannada text. On XP I do have Kannada Opentype 
> fonts.

I haven't used yudit on windows, but on Linux and BSD, I have to explicitly
point it to the TTF file via .yuditrc. It doesn't use operating system
provided facilities to deal with fonts - it deals with them directly.

Gasper wasn't keep on open type fonts last I checked.

	-Arun

Re: Letter frequency in unicode (Was Re: [Indic-computing-devel] Free UCS outline font)

[Arun S. <ar...@sh...>]

On Fri, Mar 29, 2002 at 01:26:06PM +0530, Dr. U.B. Pavanaja wrote:
> 
> People at Prajavani (a leading Kannada daily) have done the frequency 
> analysis for Kannada letters looooooooong ago. They had even designed 
> their Montype Kannada layout based on these data.

Did they publish their findings somewhere ?

> Now all these have 
> gone due to the advent of GoK (Govt of Karnataka) standard Kannada 
> keyboard layout (also know as KGP (Kannada Ganaka Parishat) std).

Did the KGP keyboard take into account such analyses ? I'm sure it's
easy to use for people already comfortable with qwerty.

If the world didn't have qwerty or English and you were (the Indian)
Sholes designing a keyboard for kannada, how would you design it ?

> 
> I hope everyone remembers the story about QWERTY keyboard that we use 
> for English. It was designed to make typing SLOW rather than FAST ;-)

Qwerty isn't the unchallenged king of the keyboard universe. I recently
bumped into Dvorak:

http://www.mwbrooks.com/dvorak/

	-Arun

Re: [Indic-computing-devel] Yudit 2.5.4

[Dr. U.B. P. <pav...@vi...>]

Hi,

If download it for Windows, can I get Unicode Kannada editor on 
Win98? What about the problem of Win98 NOT supporting OpenType fonts? 
Has it got a rendering engine for Indic scripts for Opentype fonts?

I did actually download it for Windows and tried both on Win98 and 
WinXP. I selected Kannada. Entered some unicode values for Kannada. 
But I did not see Kannada text. On XP I do have Kannada Opentype 
fonts.

-Pavanaja

> Hi,
> 
> The current version of Yudit has complete support for malayalam and
> other indic languages. It can also use Opentype layout tables of
> Malayalam fonts. I think Yudit is the first Application that can use
> Opentype tables for Malayalam, as MS is yet to release it's engine for
> malayalam.
> 
> raj
> 
> ---------- Forwarded message ----------
> Date: Sun, 10 Mar 2002 10:59:41 +0900 (JST)
> From: Gaspar Sinai <gs...@yu...>
> To: lin...@nl...
> Subject: Yudit 2.5.4
> 
> Yudit 2.5.4 has been released. It can be downloaded from:
> 
>   http://www.yudit.org/download.html
> 
> Changes:
> o Malayalam,Kannada and Telugu support
> o Software glyph-mirroring
> o iso-8859-15, iso-8859-16, koi8-c, koi8-u, ncr and rovas
>   converters
> o Ukrainian kmap and menu translations
> o Old Hungarian (rovAsA-rAs) has bee added using Unicode
>   Private Use Area: http://www.yudit.org/download/pua
> o Fallback font (yudit.ttf) encoding changed from cp-1251
>   to unicode
> o HOWTO-rovasiras.txt HOWTO-malayalam.txt has been added
> o Some of the bugs have been fixed
> 
> There are a lot more fixes that I planned but I did not
> have time for...
> Enjoy,
> gaspar
> 
> 


-----------------------------------------------------
Dr. U.B. Pavanaja
Editor, Vishva Kannada
World's first Internet magazine in Kannada
http://www.vishvakannada.com/

Note: I don't worry about pselling mixtakes