From - Mon Jul 21 21:10:28 1997 Received: from sturgeon.cs.washington.edu (128.95.2.58) by ns.cnet-sc.ne.jp (EMWAC SMTPRS 0.81) with SMTP id ; Mon, 21 Jul 1997 04:30:13 +0900 Return-Path: Received: (chou@localhost) by sturgeon.cs.washington.edu (8.8.5+CS/7.2ws+) id MAA04093; Sun, 20 Jul 1997 12:32:17 -0700 Date: Sun, 20 Jul 1997 12:32:16 -0700 (PDT) From: Pai Chou To: •@ Ëi Subject: Re: I won't to know your TW-MN-Rpmanization writing system's name. In-Reply-To: <33D208C8.4EAD108@po.cnet-sc.ne.jp> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Transfer-Encoding: QUOTED-PRINTABLE X-Mozilla-Status: 0011 Hello! Well, my tone notation uses eight (actually nine, or even ten) tone=20 characters: symbol tone# [-] 1 [=3D] 7 [.] 3 [\] 2 [/] 5 [,] 4 ['] 8 also reserved: [^] and [_] I reserve [^] for the liaised tone, which is the combination of [=3D] going into [-]. For example, "yesterday" [^zang], which is the combination of [=3Dza-heng]. "woman" [^zao/lang], which comes from [=3Dza-vo/lang]. Also= , it happens in Japanese loan words, like "engine" [^en\rin], "white shirt" [^wai'sia,zy]. Currently, Church Romanization cannot express this kind of tone, because it is relatively modern. People write it as tone 5, and I don't think it is correct.=20 My convention is to write "surface tones" (what you hear and speak),=20 rather than "lexical tones" (dictionary form, like Church Roman does). =20 It is possible to have both together! For example, [-tni -=3Do -o, ,\veu '.leo =3Dho] "sky dark dark, about to rain" where [-=3Do] means the dictionary form is [-o], but pronounced [=3Do]. However, it can become somewhat difficult to read. You are welcome to=20 suggest better ways of using it! For sound, I have more vowels than most other Romanizations. My goal is to write one form to unify dialectal differences! At the same time, I will also capture sound that other forms cannot capture. For example, in Loggang accent, Word My notation Church form "mile" [\li] li2 "you" [\ly] li2 -- cannot distinguish! "pig" [-dy] ti1, tu1 -- cannot unify! also cannot express ty1 "horse" [\vei] be2 "to buy" [\vwe] be2, boe2 -- cannot unify! if written as be2, can't distinguish from hors= e. also cannot express [b + o + eu] accent, where eu is the schwa "tail" [\veu] be2, boe2 -- cannot unify! (same problem as bu= y) also cannot express [b + eu] accent, where eu is the schwa. "to do" [.zoe] cho3, choe3 -- cannot unify! Church Roman's approach is to take TaiLam accent as standard, ignore=20 Loggang form entirely (no [y] or [eu]), and makes no attempt to unify=20 accent variations. As a phonetic notation, it is good to have accent=20 variations; but as a *writing system*, it is not good, because there is=20 no standardized form. It is not easy, but so far I have covered a big set: My form Church expansion [y] {i}, {u}, AND {y} [eu] {e}, {oe}, AND {eu} [e] {e} and {eu} (example: "to sit" [=3Dze], expand to {che7} and {ch + [eu] 7} [oe] {o} and {oe} [oei] {e} and {oe} [we] {e}, {oe}, and {u + [eu]} [eo] {o} [o] {ou} [ei] {e} I am not the first person to try this. Even Church Roman itself does a=20 little bit of unification: Church standard expansion {j} {j}, {l} {ian} {ien}, {en} {eng} {ieng}, {ing} But Liim Keahiiong's system takes one step further Liim standard expansion {ue} {e}, {oe} But too bad they only stopped there. There is no provision for my [y] or= =20 [eu]. For consonants, I use a more pinyin like approach, because I find young=20 people can accept the one-letter-one-sound assignment more easily. It=20 has less conflict with their English intuition. Anyway. My ambition is to make a computer dictionary in this form, so=20 that it contains the most completion information. From this common form,= =20 I can use a computer program expand each word into the variants, and=20 convert automatically between different Romanizations, so that many users= =20 can use it. Right now there is too much duplicated effort. I was wondering if it is possible for me to get Dr. Ong's dictionary file? Currently I have an experimental dictionary of 3200 words but it just=20 contains random entries. It will definitely help me! thanks, Pai On Sun, 20 Jul 1997, =C9=A1 =CBi wrote: > Hai. > Your writing system =3D*'*-* etc. is very easy to typing on note. > Now I try to make WWW TaiLamDictionary from=20 > Dr Ong_Iok-Tek 's dictionary's 5000 words. > But I don't know that how to write TW-MN-kana on PC. > And I don't know how to write these tones. > I hope to use your writing-system for the TaiLam5000WordsDict's tone. > Please teach me you system's name, and usage, etc. > chhiaN2 li2 chi2-kau3. >=20 > Itoh Shohji. > benzo@po.cnet-sc.ne.jp > http://www.cnet-sc.ne.jp/benzo >=20 > I hope to know where is the plase to looking for your new URL-address. >=20