I get this:To summarize, for tamizh #n/#ng is ங், and "ng" is ஙக்
For kannada (and similar), #n/#ng is ಙ್, and "ng" is Mg (i.e. anuswara)
Done. btw i added explicit support for tamizh ந by using ^n. i know you liked 2, but i thought ^nAmam is less intrusive than n2Amam - so i wanted people to get a feel for it. Something other than ^ can work too if people dont like it. Its easy to change the logic on this to another symbol. Of course, ^n and n are the same for kannada (ನ)arunk wrote:ok thanks. I will upload this version for testing. Thanks!drshrikaanth wrote:So far ok.
. But 2n is not that great and probably can cause other problems. . I am not overly married to ^ in particular, but i prefer it better than n2. in tamizh (you might use S or Sh or whatever - as long as it gives consistent results in both scripts - AND - you tell the user beforehand how he should spell this sound Before that another observation. The anunAsika (ma kAra) before a avargIya vyanjana should be represented by a anusvAra in kannaDa.ramakriya wrote:more about ~nAna and saMskR.ta in the next post.
-Ramakriya
Try "S" for this. Both "sh" and "Sh" stands for the next one (i.e. ಷ ஷ).ramakriya wrote:sh - should be ಶ in kannada, and ச, as for as I know
Ok. I will check this.ramakriya wrote:The samyuktAkshara 'ksha' is rendered correctly in tamizh script, but incorrectly in kannada
Based on the representation for 3 sa's are "sa", "Sa" and "Sha/sha": i presume this is always ksha (or is there kSa too which results in a single letter)?(kSa gives the correct representation in kannaDa script ( ಕ್ಷ ) but wrongly shows up as க்ச in tamizh. kSha or ksha give the correct representation in tamizh ( க்ஷ ) but shows up as ಕ್ಶ in kannaDa.
Sorry. I dont know what those terms mean. Can you explain it in simpler terms please?ramakriya wrote:Before that another observation. The anunAsika (ma kAra) before a avargIya vyanjana should be represented by a anusvAra in kannaDa.ramakriya wrote:more about ~nAna and saMskR.ta in the next post.
-Ramakriya
Here is what I get:
samyOjane samhAra samvahana samShOdhane samsAra
சம்யோஜனெ சம்ஹார சம்வஹன சம்ஷோதனெ சம்சார
ಸಮ್ಯೋಜನೆ ಸಮ್ಹಾರ ಸಮ್ವಹನ ಸಮ್ಶೋಧನೆ ಸಮ್ಸಾರ
-Ramakriya
You have spelt j~nAna wrongly. The jakAra must occur first and the anunAsika vyanjana is only a ottakShara.ramakriya wrote:arunk,
Tried out the following text in the test bed for testing s (as in sangam), sh (shankara) and Sh (Shanmukha) along with the visarga feature you added;
karuNA kaTAkSha Ela upEkSha duShTa shikShaka shiShTa rakShaka - duHkhahara shiva kR.pALu susamskR.ta bhaktara poreyO ~nAna dIkSe ittu.
In kannada it shows up as:
ಕರುಣಾ ಕಟಾಕ್ಶ ಏಲ ಉಪೇಕ್ಶ ದುಶ್ಟ ಶಿಕ್ಶಕ ಶಿಶ್ಟ ರಕ್ಶಕ - ದುಃಖಹರ ಶಿವ ಕೃಪಾಳು ಸುಸಮ್ಸ್ಕೃತ ಭಕ್ತರ ಪೊರೆಯೋ ಙಾನ ದೀಕ್ಷೆ ಇತ್ತು.
This is not clear and not always the case either. You have words like saumya, and Amla and namrate, where the anuswAra/bindu is Not used. Here, the bindu occurs only when "sam" or some other such prefix is present which is not an integral part of the word e.g saMyAna, saMlApa, saMrAT etc.ramakriya wrote:Before that another observation. The anunAsika (ma kAra) before a avargIya vyanjana should be represented by a anusvAra in kannaDa.ramakriya wrote:more about ~nAna and saMskR.ta in the next post.
-Ramakriya
you are right about ya, ra and la - Did not think of these casesdrshrikaanth wrote:This is not clear and not always the case either. You have words like saumya, and Amla and namrate, where the anuswAra/bindu is Not used. Here, the bindu occurs only when "sam" or some other such prefix is present which is not an integral part of the word e.g saMyAna, saMlApa, saMrAT etc.ramakriya wrote:Before that another observation. The anunAsika (ma kAra) before a avargIya vyanjana should be represented by a anusvAra in kannaDa.ramakriya wrote:more about ~nAna and saMskR.ta in the next post.
-Ramakriya
With v, S, Sh and s and h, I think the bindu always occurs regardless of the form of the word. So the first 3 consonants of the avargIya vyanjanas(those that are not grouped as pentads) allow for both usages while the rest alwyas have the anuswAra when m is the part of a conjunct consonant.
I can incorporate this if needed, so that you can get away with samh... or saMh... etc. If it is ambiguous only for "y", "r" and "l", i can make it so for any other consonant. That way an explicit bindu specification would be needed only if 'ma' consonant is followed by ya, ra or la variant. Would that be ok?With v, S, Sh and s and h, I think the bindu always occurs regardless of the form of the word. So the first 3 consonants of the avargIya vyanjanas(those that are not grouped as pentads) allow for both usages while the rest alwyas have the anuswAra when m is the part of a conjunct consonant.
Arun. Why give room for confusion. Simply male M and m no interchangeable. Tomorrow another exception could pop up tha would mandate this anyway.arunk wrote:I can incorporate this if needed, so that you can get away with samh... or saMh... etc. If it is ambiguous only for "y", "r" and "l", i can make it so for any other consonant. That way an explicit bindu specification would be needed only if 'ma' consonant is followed by ya, ra or la variant. Would that be ok?
I dont think j~nAna would confuse anyone. That is the spelling I have seen used almost everywhere ever since my schooldays. This is alsothe way it is spelt in writing in kannaDa. In case you are not aware, the pronunciation of j~nAna in kannaDa and ~nAna in tamizh are different. This is how it looks ಜ್ಞಾನ. Can you see?For ??? , it is ~nAna or ~nyAna (i thought i mentioned it before). j~n will confuse it presently as ramakriya observed. Anyway, ~nAna/~nyAna now comes out as ???. I am presuming now that this is wrong. How should it be?
ok. if we are not sure that we have all cases - no need to make M/m interchangeable.drshrikaanth wrote:Arun. Why give room for confusion. Simply male M and m no interchangeable. Tomorrow another exception could pop up tha would mandate this anyway.
. Now i do remember the kannada spelling. I need to go back and check what the logic is doing - i dont think i incorporated it. I will fix this later today. I will make j~n/j~ny <=> ~n/~ny for tamizh. BTW, do telugu and sanskrit treat it like kannada (so that i can reuse the kannada logic as is)Yes you could do that.arunk wrote:Atleast, we should make "m/M" interchangeable at the end of a word.
Iam not sure what you mean here but if you are writing sanskrit in sanskrit, the "m" at the end also will be "m" only and not the circle/bindu.That will fix the above case and can even make many sanskrit krithis easier with many words ending in "m" for dvitIya vibhakti (although in such cases, i would think most people notating sanskrit krithis would know sanskrit and so when to use M and when not).
ANother thing I probably did not mention earlier. When m combines with itself as a conjunct as in amma, it is Never shown by bindu. It is alwas amma i.e as the latter itself.The same argument can apply for "m" in the middle for tamizh krithis also. I dont yet know for sure as what to do there but the range of consonant sounds that can follow mei 'm' (ma) in tamizh words (native ones - see below) is much more limited i think. I can think of"sa" (imsai), "ba" (ambu), "ma" ( amma)? Perhaps these can morph to 'M' if approp? But that would apply only if such combinations always require bindu in kannada/telugu/sanskrit etc. Based on what you said before, i think this may be the case (??).
It is vA#nmaya. Not vA#ngamaya. And the pronunciation is perceptible and clearly ddifferent from vAMgAdE.arunk wrote:The question is the actual phonetic sound of ಜ್ಞಾನ is (even slighltly) different from the tamizh word ஞான (as in a tamizh krithi), i.e. as sort of how vA#ngamaya is different in saMgam (or vAngadE => vAMgAdE)?
I think you are safer writing it as j~nAna in kannaDa as that is how kannaDigas will pronounce it even in a tamizh kRti. The problem with lwriting it as ~nAnam is that they will probably have problems recognising the sound as words never begin with ~n in kannaDa.If so then if i have want to transliterate ஞான in a tamizh krithi as ~nyAna, how should it appear in kannada? Note that going other way is fine j~nya/j~na => ஞ
Iam afraid you cannot have the good news. There are many words such as aj~na, sarvaj~na, aj~nAna, saMj~ne etc in kannaDa.Also i presume j + ~na/~nya in kannada is not a special case and that there may be other combination of "some consonant" + ~na/~nya? If not that would be good news
? So bindu for "m" at end of word is for kannada and not for sanskrit or is it not always the case for sanskrit? Ramakriya for bhajarE had:drshrikaanth wrote:Iam not sure what you mean here but if you are writing sanskrit in sanskrit, the "m" at the end also will be "m" only and not the circle/bindu.
?Both are/should-be equiv. in this scheme currently (only because #ng is phonetically closer than #n)drshrikaanth wrote:It is vA#nmaya. Not vA#ngamaya.
Yes i did see that.And the pronunciation is perceptible and clearly ddifferent from vAMgAdE.
No i meant if word occurs in tamizh krithis, i would like to avoid saying it as j~nAna.I think you are safer writing it as j~nAna in kannaDa
The interpretor (logic) can do it so it interprets it like jn~ (which is why i asked that the question below)The problem with lwriting it as ~nAnam is that they will probably have problems recognising the sound as words never begin with ~n in kannaDa.
No I meant some thing other than "j" before ~n. If it cannot, then j~n <=> j~ny <=> ~n <=> ~ny always and we have no problems. You can say ~nyAna/~nAna in a tamizh krithi, j~nAna/j~nyAna in a kannada krithi and it will come out right in both languages (with the help of the additiional logic). But if ~n can occur without j before it, that would be a problem.drshrikaanth wrote:Iam afraid you cannot have the good news. There are many words such as aj~na, sarvaj~na, aj~nAna, saMj~ne etc in kannaDa.Also i presume j + ~na/~nya in kannada is not a special case and that there may be other combination of "some consonant" + ~na/~nya? If not that would be good news
Ok. This should be no problem. The sanskrit interpretor can treat m/M as "m".drshrikaanth wrote:No Arun. AFAIK the letter m only occurs at the end in sanskrit, not the bindu. In hindi of course it is different. But people often might simply follow the pattern in their language while writing sanskrit as well *and hence the M or the dot at the end). But strictly, it should be the letter only.
I do have an issue with the latter spelling vA#ngmaya. For those who read the english version, it introduces an exztra "g" sound which should not be there.arunk wrote:Both are/should-be equiv. in this scheme currently (only because #ng is phonetically closer than #n)drshrikaanth wrote:It is vA#nmaya. Not vA#ngamaya.
I did get what you meant. Iam saying there will be some peculiarities of pronunciation by non-native speakers. No doubt some who are familiar with tamzih pronunciation will get it right but most will pronounce it as j~na only.arunk wrote:No i meant if word occurs in tamizh krithis, i would like to avoid saying it as j~nAna.drshrikaanth wrote:I think you are safer writing it as j~nAna in kannaDa
I meant they might simply not able to discern how to pronounce if you just wrote ~jAnam. Is it not better to at least be able to sing albeit with imperfect diction. If you want you can leave it as ~na rather than j~na for tamizh krtis.arunk wrote:The interpretor (logic) can do it so it interprets it like jn~ (which is why i asked that the question below)drshrikaanth wrote:]The problem with lwriting it as ~nAnam is that they will probably have problems recognising the sound as words never begin with ~n in kannaDa.
hmm. No, I cannot think of other combinations offhand.No I meant some thing other than "j" before ~n. If it cannot, then j~n <=> j~ny <=> ~n <=> ~ny always and we have no problems. You can say ~nyAna/~nAna in a tamizh krithi, j~nAna/j~nyAna in a kannada krithi and it will come out right in both languages (with the help of the additiional logic). But if ~n can occur without j before it, that would be a problem.
Sorry. I know you got what i meant, but i wanted to i would avoid having to write it in the scheme as j~nAna for a tamizh krithi. You are right that pronounciation will invariably vary among different speakers. While we can "help" some with super-script suffixes etc., you can only go so far.drshrikaanth wrote:I did get what you meant. Iam saying there will be some peculiarities of pronunciation by non-native speakers. No doubt some who are familiar with tamzih pronunciation will get it right but most will pronounce it as j~na only.arunk wrote:No i meant if word occurs in tamizh krithis, i would like to avoid saying it as j~nAna.
Ok. For now let me go with this.drshrikaanth wrote:hmm. No, I cannot think of other combinations offhand.No I meant some thing other than "j" before ~n. If it cannot, then j~n <=> j~ny <=> ~n <=> ~ny always and we have no problems. You can say ~nyAna/~nAna in a tamizh krithi, j~nAna/j~nyAna in a kannada krithi and it will come out right in both languages (with the help of the additiional logic). But if ~n can occur without j before it, that would be a problem.
You have a few words in sanskrit where ~n occurs with c andjust 2 or 3 with l (l~n God knows how to pronounce it). I must say these are very unlikely to occur in kRtis. The l~n combination- never. c~n possible but very slim chance.arunk wrote:No I meant some thing other than "j" before ~n. If it cannot, then j~n <=> j~ny <=> ~n <=> ~ny always and we have no problems. You can say ~nyAna/~nAna in a tamizh krithi, j~nAna/j~nyAna in a kannada krithi and it will come out right in both languages (with the help of the additiional logic). But if ~n can occur without j before it, that would be a problem.
Ok. If you think #ng will mislead people, i can leave it as #n. I think usage of #n (when compared to ng/Mg) is rarer in tamizh as well as kannada . Just to complete the circle, i would like a malayalam perspective where i think this consonant is used a lot more.drshrikaanth wrote:I do have an issue with the latter spelling vA#ngmaya. For those who read the english version, it introduces an exztra "g" sound which should not be there.arunk wrote:Both are/should-be equiv. in this scheme currently (only because #ng is phonetically closer than #n)drshrikaanth wrote:It is vA#nmaya. Not vA#ngamaya.
Even here there is hope. We can always first look for "j~n", "l~n" and "c~n" as a single unit, and if not then "~n" becomes "j~n".drshrikaanth wrote:You have a few words in sanskrit where ~n occurs with c andjust 2 or 3 with l (l~n God knows how to pronounce it). I must say these are very unlikely to occur in kRtis. The l~n combination- never. c~n possible but very slim chance.arunk wrote:No I meant some thing other than "j" before ~n. If it cannot, then j~n <=> j~ny <=> ~n <=> ~ny always and we have no problems. You can say ~nyAna/~nAna in a tamizh krithi, j~nAna/j~nyAna in a kannada krithi and it will come out right in both languages (with the help of the additiional logic). But if ~n can occur without j before it, that would be a problem.
Sorry to butt in Arun. But from the little I know and the lot I have heard, there is no place for g in pronunciation(and certainly in spelling as you know).arunk wrote:jayaram - the question is Does the malayalam consonant closest to "#n" (as in vA#nmaya, a#n#anam/அங்ஙனம்), better phonetically represented (for malayalam words only) by #ng or would it mislead people by overemphasizing the "ga" sound"?
ok. I will remove it. We can add it if many people later feel like it makes sense in some contexts (and even then we can recommend against it for kannada, sanskrit etc.)drshrikaanth wrote:Sorry to butt in Arun. But from the little I know and the lot I have heard, there is no place for g in pronunciation(and certainly in spelling as you know).arunk wrote:jayaram - the question is Does the malayalam consonant closest to "#n" (as in vA#nmaya, a#n#anam/அங்ஙனம்), better phonetically represented (for malayalam words only) by #ng or would it mislead people by overemphasizing the "ga" sound"?
)DRS is absolutely correct on this - For bhajarE rE, the transliteration scheme used was for kannDa - I used the same text to generate samskrita unicode - AFAIK samskrita would use half consonent here.arunk wrote:? So bindu for "m" at end of word is for kannada and not for sanskrit or is it not always the case for sanskrit? Ramakriya for bhajarE had:
bhajarE rE mAnasa mahiShAsura mardineeM ||
nija bhakta jaya shubhadAyineeM ||
bhavasAgara tAraka roopAM ||
shree vidyAM mahA kAma suMdareeM ||
From what little I know the sanskrit "rendition" of it do have dots above (just like kannada as anuswara following). In any case, i was thinking it would be easier to get this right for languages that require "ma" at end to require bindu/anuswara. Am i confused again
Arun
That's interesting - do you know any of these words?drshrikaanth wrote:You have a few words in sanskrit where ~n occurs with c andjust 2 or 3 with l (l~n God knows how to pronounce it). I must say these are very unlikely to occur in kRtis. The l~n combination- never. c~n possible but very slim chance.
If I were to write a tamizh kriti with the word anbu - I would retain it as such. 'nb' here should not be replaced with 'mb' .arunk wrote:I am adding support for M etc. and i have a question. How do the tamizh words "anbu" (affection) and "ambu" (arrow), get written in kannada. Or more generally does "nb" become "Mb"? (and if so, does "mb" remain "mb')?
Arun
yAc~nA and other combination words with yAc~na, pUgayal~naramakriya wrote:That's interesting - do you know any of these words?drshrikaanth wrote:You have a few words in sanskrit where ~n occurs with c andjust 2 or 3 with l (l~n God knows how to pronounce it). I must say these are very unlikely to occur in kRtis. The l~n combination- never. c~n possible but very slim chance.
Ramakriya has already answered this. To elaborate, mb uses bindu only because m and b both belong to the same pentad. Likewise. all combinations where the first nasal consonant and the 2nd consonant of a conjunct beong to the same pentad, the bindu will occur. Combinations of a nasal consonant with a consonant from another varga/pentad will retain the nasal as such. bindu will not/cannot occur here.arunk wrote:I am adding support for M etc. and i have a question. How do the tamizh words "anbu" (affection) and "ambu" (arrow), get written in kannada. Or more generally does "nb" become "Mb"? (and if so, does "mb" remain "mb')?
ARun read my previous post on the rules for M occurring in some places.arunk wrote:Right now in ng, np, nc, nT, nt, the 'n' becomes 'M'. So 'nb' would remain as 'nb'
FinePer new rules:;
1. M is explicit bindu. It translates to "ma" for tamizh only.
2. Either m or M at end becomes bindu for kannada.
These words in tamizh are anyway borrowings from sanskrit.3. 'ms' and 'mb' also become 'Ms' and 'Mb' to account for tamizh (and is consistent with kannada).
ramya(again a sanskrit word) will be ramya. But saMyama will have bindu. You did not forget what I wrote about both m and M occuring in combination with y, r and l did you?4. The other tamizh occurences my (ramya), mm (amma) wont become M in kannada, and that is consistent.
. I see the pentats are:)? Would these considered nasal consonants that follow these rules? (so kaNDEn in tamizh would use bindu? No i didnt. I am going to leave user to specify bindu here (i.e. "my" would remain "my", and if bindu applies you have to specify as "My"). For most tamizh words that you use "my" combo (yes - i do agree that they will be sanskrit imports), i am hoping/guessing they are not required. But it turns out to be so for some of them, user can specify it. They will be rare anyway in tamizh krithis, and asking for explicit qualification there is ok.drshrikaanth wrote:ramya(again a sanskrit word) will be ramya. But saMyama will have bindu. You did not forget what I wrote about both m and M occuring in combination with y, r and l did you?4. The other tamizh occurences my (ramya), mm (amma) wont become M in kannada, and that is consistent.
Now that is good newsdrshrikaanth wrote:And finally, when you introduce teugu, you can safely apply the same rules as for kannaDa as is.
First you are confusing us all by using differen representations for some of the letters (nga, nya)arunk wrote:Silly questions. Please be patient
k, kh, g, gh, nga,
ca, cha, ja, jha, nya,
Ta, Tha, Da, Dha, Na,
ta, tha, da, dha, na,
pa, pha, ba, bha, ma.
Already answered above.Also do these rules apply for Na too (nga? nya? doesnt seem so but what do i know
yes. So also in kannaDa (kaMDe)(so kaNDEn in tamizh would use bindu?
that gives it wrong as well (first anunAsika in ottu instead of the second)