Re: [m17n-list] Questions about hi-itrans.mim

[विश्वासो वासुकेयः / Vis hvas Vasuki]

On Sun, Oct 28, 2012 at 3:02 AM, Kenichi Handa <address@hidden> wrote:

Thank you for the web pages.  I have several questions for the transliteration table...

(1) Are you proposing the new name "OptiTrans"?

Not yet - I anticipate that the suggestions described there may evolve to a point where there might be the need to make a significant break from ITRANS and submit different *.mim files. (Due to the fact that many presses of the key 'a' can be avoided without sacrificing intuitiveness.) But that is for later (if it works out in tests). 

So for now, it might suffice to change the description to include the following:

"The motivation behind additions made to the basic ITRANS scheme is described in http://sanskritnlp.appspot.com/optitrans.html , along with a tabulated comparison with several other transliteration schemes."

 

(2) What does ".\1" mean? (for क़, ख़, ग़, etc.)

I was using a half-baked regular _expression_ language. 

.\1 means that '.' followed by the character sequence (eg: k, kh) should be mapped to the nukta character followed by the corresponding devanAgarI character (क्, ख्).

I wish there were a more beautiful way of expressing it - The alternative - .k .kh ... to क़, ख़ ... seems lacking in completeness.

 

(3) What does "n\1" mean?

Just as the above, 

 

(4) Do you want all bold keys implemented in hi-itrans?

Yes. Your expertise would be very useful with the most tricky maps - like
a] .<whitespace> = ।<whitespace>

b]  ./ = forget context. Eg: tita./u =  तितउ (not तितौ) and k./h = क्ह् (not ख्).

--
Vishvas / विश्वासः