# Tamil - Morse-code #170

Closed
opened this Issue Aug 15, 2018 · 5 comments

Projects
None yet
2 participants
Collaborator

### arcturusannamalai commented Aug 15, 2018

 Can we compose a Tamil Morse-code ? Start with a frequency count of Tamil letters from various sources Build a probability distribution from the frequency counts Build a Huffman code using the above distribution Each letter of Tamil alphabet gets a Morse code : 0 = '.', 1 - '-'. புள்ளி, கோடு.

Collaborator

### arcturusannamalai commented Aug 15, 2018

 commit ac9e9dc

Collaborator

Contributor

### ssurenr commented Aug 16, 2018

 Isn't 30 symbols enough?
Collaborator

### arcturusannamalai commented Aug 17, 2018

 I think you are saying ceil(log2(30)) = 5bits is sufficient, by virtue of Abugida / Uyir-Mei syllabic nature of Tamil. However when expressed in code, it may require some complex decoding logic. This is a contentious matter! Ithanai sariyaaga seivatharkku comment-section-ai thaandi muyarchikka vendum; yenn ninga ithanai oru script-aaga meroklla koodathu @ssurenr ? tamil.utf8.splitMeiUyir can help!
Contributor

### ssurenr commented Aug 19, 2018

 ஆம். 247 மோர்ஸ் சிம்பல்களை நினைவில் வைப்பதை விட 30 எளிதல்லவா. டீகோட் செய்யும்போது வரும் சிக்கலுக்கு ஒரு எடுத்துக்காட்டு தர இயலுமா? ஒரு issue உருவாக்கி எனக்கு ஒப்படையுங்கள்.