Tarquin saith:
Everyone tries to invent their own phonetic system when they're in their teens. I did. But reinventing the wheel is generally a bad idea. if every dictionary & encyclopedia has its own system, then it is not transferable, and the reader has to relearn for each book they open. Let's stick to something that is universal: IPA / SAMPA.
I've seen IPA/SAMPA used in Wikipedia articles and didn't like it. The former uses too many non-ASCII characters to make it easy enough to use and the latter uses weird ASCII signs that make the word @n"rid@bl [unreadable] and l33t-like for those who don't know the system.
The system used in practically every school dictionary I've seen (except for Merriam-Webster's) is simple: prime mark after accented syllable (doubled for two accents in one word), flipped e for schwa, breve (curve) for short vowels, macron (line) for long vowels, diaresis or circumflex for vowels of far and fur, line across th for voiced and no line for unvoiced. This last one is the only thing not representable in Unicode; everything else is sĭm'pəl ēnəf' for us to use. (If those words came out garbled, either your mail client or mine doesn't support Unicode. That was 73 12D 6D 27 70 259 6C 20 113 6E 259 66 27 hex.)
Even better is the limited letters-only system used in textbooks that aren't dictionaries: complex words are repeated in identical-sounding imitations (I-mi-TAY-shuns). If this latter system is formalized (FAR-muh-liyzd) it may be simpler than SAMPA for those who don't have the time to learn it.
Geoffrey Thomas / jěf'rē t�'məs / JEF-ree TAH-muhs geoffreyerffoeg@yahoo.com
__________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com
Geoffrey Thomas wrote:
line across th for voiced and no line for unvoiced. This last one is the only thing not representable in Unicode;
Not so. Even 8859-1 makes this accomodation because of its need to be compatible with a modern Western language, i.e. Icelandic. Thus #0240 is "ð" (voiced) and #0254 is "þ" (unvoiced). To enter these in an e-mail I simply type the number (with its leading zero) while holding the Alt key.
Ec
Take a look at my box and see what you think.
http://www.wikipedia.org/wiki/User:Stevertigo/Sand_box
Im finding these edit tags poping up everywhere is noying as heck. I just turned em off in the prefs -- maybe they should default to off, and then maybe a __NOEDTAG__ feature for special cases might be appropriate.
-S-
__________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com
steve-
Take a look at my box and see what you think.
This highlights a bug in the parser. They should really not be visible on that page at all.
Regards,
Erik
steve-
Take a look at my box and see what you think.
Correction - it's not a bug. You are using <h1> where it shouldn't be used. Don't do that.
Regards,
Erik
Erik said: Correction - it's not a bug. You are using <h1> where it shouldn't be used. Don't do that.
So - the <BIG> tag wont trigger the header -- OK that seems to work fine, then. Danke mucho.
What about the notion of a _NOSECEDIT__ tag? And what about a text margin on the right side so text doesnt merge with the [EDIT] tag? A bit more margin there regardless of tags might be easier on the eye.
'Gards -S-
__________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com
While Tarquin's comment was a bit cheeky (what use would he have had *then for it) -- and Dan's comment earlier about professors knowing better than the rest of us, so why bother (well?) -- represent the pinnacle of interest in the subject... Still -- wikipedia might have an interest in attempting to unify/formalize something on this. Maybe its time that
http://meta.wikipedia.org/wiki/Phonetic_schema
But the whole notion of using English specific formats is useless-- once again the see we tend to swim in is an English-centered one.
And Ill take a look at any PHP or Perl pinyin converters so that stuff like "pin:/wo3 bao4 you3/ SARS" gets turned in into properly accented vowels, etc.
__________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com
Stevertigo wrote in part:
But the whole notion of using English specific formats is useless-- once again the sea we tend to swim in is an English-centered one.
OTC, it's /exactly/ what's useful for English words.
As I said when I first brought up morphophones, a morphophonic transcription should only be used for English words, which appear (or may appear) in all dialects, with different phonemes. Foreign proper names should still be rendered phonemically, according to the way that the relevant foreigners speak.
-- Toby
steve vertigo wrote:
While Tarquin's comment was a bit cheeky (what use would he have had *then for it) -- and Dan's comment earlier about professors knowing better than the rest of us, so why bother (well?) -- represent the pinnacle of interest in the subject... Still -- wikipedia might have an interest in attempting to unify/formalize something on this. Maybe its time that
http://meta.wikipedia.org/wiki/Phonetic_schema
But the whole notion of using English specific formats is useless-- once again the see we tend to swim in is an English-centered one.
And Ill take a look at any PHP or Perl pinyin converters so that stuff like "pin:/wo3 bao4 you3/ SARS" gets turned in into properly accented vowels, etc.
How about something similar to the ISBN links that generates SABLE/MS SAPI markup/whatever the heck Apple use &c?
Geoffrey Thomas wrote:
I've seen IPA/SAMPA used in Wikipedia articles and didn't like it. The former uses too many non-ASCII characters to make it easy enough to use
Currently, that's true, but technology is evolving ... ;-)
and the latter uses weird ASCII signs that make the word [unreadable] and l33t-like for those who don't know the system.
I agree to that. ASCII is not suitable for phonetic representation.
The system used in practically every school dictionary I've seen [...] everything else is sĭm'pəl ēnəf' for us to use. (If those words came out garbled, either your mail client or mine doesn't support Unicode. That was 73 12D 6D 27 70 259 6C 20 113 6E 259 66 27 hex.)
You've sent HTML entities in a plain-text mail. ;-) No compliant mail/news client should interpret them.
Here's what you were trying to write: everything else is sĭm'pəl ēnəf' for us to use. Geoffrey Thomas / jěf'rē tä'məs / JEF-ree TAH-muhs
A Unicode-capable mail/news client should be able to display this correctly.
However, I'm not a great fan of this system either. It is not significantly easier to generate than IPA, and it is certainly less universal (I, for once, have never come across it; maybe it's Americentric).
(I-mi-TAY-shuns). If this latter system is formalized (FAR-muh-liyzd) it may be simpler than SAMPA for those who don't have the time to learn it.
I always thought this system looked really un-pro-FESH-un-al.
Greetings, Timwi
Timwi wrote:
Here's what you were trying to write: everything else is sĭm'pəl ēnəf' for us to use. Geoffrey Thomas / jěf'rē tä'məs / JEF-ree TAH-muhs
A Unicode-capable mail/news client should be able to display this correctly.
However, I'm not a great fan of this system either. It is not significantly easier to generate than IPA, and it is certainly less universal (I, for once, have never come across it; maybe it's Americentric).
I thought it was fairly universal myself -- it's in every major dictionary I've used, including some of the results at www.dictionary.com. It's not particularly easy to generate, but it seems much easier to *read*.
-Mark
Timwi wroteth: I always thought this system looked really un-pro-FESH-un-al.
ROFL. This *should be the absolute last word on the "American system" for pronunciation, but Delirium said:
"I thought it was fairly universal myself -- it's in every major dictionary I've used, including some of the results at www.dictionary.com. "
Well- the point was that this way of doing it was "Americentrist" -- which Mark, you actually *validated somewhat - nothing personal. Its easy to forget the international, ie non-English thing if your not used to thinking this way.
My final thoughts on the subject before actually doing something about it: 1. SAMPA, though based on IPA was designed for machine readability -- not for human readability. 2. The notion of a human-usable phonetic schema rests on the fact that most of the world - whether its Latin, pinyin, romaji, cyrillic, and just plain non-Anglo English -- the Roman alphabet is ubiquitous and differs rather little in terms of the way its used. 3. Jimmy's point is on the right track. -- a schema for using Sampa-like input, which is then simply changed to more readable characters -- these would probably be have to be as ubiquitous as possible as well -- as long as they borrow from the better ideas out there, and could still have a 1:1 conversion to sampa (eugh!)
-S-
__________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com
PS -- its important to note that Sampa was originally designed with only Germanic /Latin languages in mind to begin with: "Danish, Dutch, English, French, German, and Italian (by 1989); later to Norwegian and Swedish (by 1992); and subsequently to Greek, Portuguese, and Spanish (1993)."
Plus it just tries to doo too damn much.
-S-
__________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com
steve vertigo wrote:
PS -- its important to note that Sampa was originally designed with only Germanic /Latin languages in mind to begin with: "Danish, Dutch, English, French, German, and Italian (by 1989); later to Norwegian and Swedish (by 1992); and subsequently to Greek, Portuguese, and Spanish (1993)."
Plus it just tries to doo too damn much.
My further objection to SAMPA is that many of its representations tend to be counterintuitive. When it uses an upper case letter to indicate something other than its minuscule counterpart I get a completely wrong impression about how the word in question should be pronounced. Although I tend to mostly favour IPA, the use of capital letters in the popular American system to indicate emphasis is very easy to understand and follow.
Ec
Timwi wrote:
Geoffrey Thomas wrote:
Geoffrey Thomas / jěf'rē tä'məs / JEF-ree TAH-muhs
Yes but for a UK speaker, the above would be wrong. TAH and TOH is not the same. (we have an article on it with sound samples but I can't find it)
Tarquin wrote:
Timwi (or Geoffrey Thomas?) wrote:
Geoffrey Thomas / JEF-ree TAH-muhs
Yes but for a UK speaker, the above would be wrong. TAH and TOH is not the same. (we have an article on it with sound samples but I can't find it)
This is /exactly/ the sort of thing that morphophones are for.
Phonemically, the short "O" may be /A/, /A<rnd>/, /O/, maybe more. Morphophonically, it's always a short "O".
A good morphophonic system will also distinguish "log" and "dog".
-- Toby
On Sat, 2 Aug 2003 15:47:20 -0700, Toby Bartels toby+wikipedia@math.ucr.edu gave utterance to the following:
A good morphophonic system will also distinguish "log" and "dog".
There are accents/dialects in which these two are diffferent?
On Tue, Jul 29, 2003 at 06:58:03PM -0700, Geoffrey Thomas wrote:
Even better is the limited letters-only system used in textbooks that aren't dictionaries: complex words are repeated in identical-sounding imitations (I-mi-TAY-shuns). If this latter system is formalized (FAR-muh-liyzd) it may be simpler than SAMPA for those who don't have the time to learn it.
That doesn't seem to bear any relation to the way I pronounce formalised, and for what it's worth I "vote" that only names should have any kind of pronunciation information; leave that to dictionaries, which can concentrate on particular dialect(s).
--- Jason Williams jason@jasonandali.org.uk wrote:
That doesn't seem to bear any relation to the way I pronounce formalised, and for what it's worth I "vote" that only names should have any kind of pronunciation information; leave that to dictionaries, which can concentrate on particular dialect(s).
Well, wiktionary at least should use a phonetic system for all words. LittleDan
__________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com