I was just reading the pronunciation of “Rodeo” thread in which Dopers spent many posts fumbling to explain to one another which exact phonetic sounds they were talking about. They found it difficult to specify exact sounds, and could convey their meaning only by a series of approximations, and even then comprehension by other Dopers was uncertain.
What impeded their understanding one another was the lack of a single shared graphic transcription system with one to one correspondence of symbol and sound. Like the IPA.
This happens all the time, mostly in GQ, whenever we have threads about pronunciations, dialects, and accents. And you know we get a lot of threads like that. I’ve seen exasperating situations develop because it’s hard to talk about phonetics without a good transcription system. I’ve seen a lot of posts wasted because of the lack of a system to convey clear phonetic information within the limitations of current internet technology.
IPA symbols are available in Unicode, Wikipedia and Wiktionary use them all the time, and Dopers use Wiki* all the time, so the concept is already pretty widely known. If some computers cannot display the IPA characters of Unicode, then we need another way to convey the same information. That’s why linguists who communicate online came up with clujes to transcribe IPA symbols using only 7-bit ASCII characters, which can be displayed the same anywhere.
One ASCII cluj designed to substitute for IPA is called X-SAMPA. I propose we adopt that as the way to talk about phonetics. It would remove so much of the noise from our signals. Another IPA-substitution system is Kirshenbaum. I propose we pick one of these, or some IPA-substitution system that works as well, and post it in a sticky in GQ as a public service.
Being an ASCII implementation is only one very small part of the problem. Transcribing into any such system, for the laymen, is at best hit-and-miss. I might know what sounds I want to describe, but have no idea where they are on that chart.
No, what “impeded” the understanding was intentional obtuseness and sardonic “humor”, as well as ignorance of how to transcribe phonetic pronounciation.
How to transcribe phonetics is a skill that can be learned. The Dope is a great place to learn new things. If anyone enters a GQ thread about phonetics, looking for information, the other Dopers there who already know phonetics will help them to get started.
With so many threads about phonetics, there is an incentive to learn and use a system that would bring a noticeable improvement to the smoothness of the discussion. Those who used it and thereby made discussions easier to follow would earn Doper Prestige Points for good netizenship. It would become one more point of excellence in the SDMB above ordinary message boards.
Yes. Look at Wikipedia’s International Phonetic Alphabet page. Each individual symbol is linked to its own page of phonetic description, with an audio file of it.
I discovered that Windows Media Player will not play nice with those audio file formats used by Wiki*. Apparently Bill Gates failed to learn anything of lasting value in kindergarten. So I downloaded Winamp to be able to play them.
Anyway, I was thinking since Dopers are constantly running to Wiki* for information and referring one another there, we’ve all seen IPA transcriptions of pronunciations. The best Wikipedia articles tend to have IPA pronunciation right at the beginning of the article. So it isn’t as unfamiliar a system to Doperdom in general as it might have been some years ago.
Merriam-Webster was developed specifically for a set of standardized American English phonemes. It lacks the symbols to transcribe different English dialects, let alone other languages. To compare across languages, we really need IPA.
The original SAMPA transcription developed during the early internet years was combined from the systems already used by several different languages–so it was adequate for covering those languages, but not necessarily others. The point of the eXtended-SAMPA cluj devised by British linguist John C. Wells is to be a unified system covering the entire range of symbols in the IPA. This does not privilege any individual languages (except insofar as it’s Roman-alphabet-based).
This is a great idea, Johanna. The X-SAMPA system is fine. Maybe we lovers of linguistics could just begin using it unilaterally. Eventually, because of its convencience and clarity, it will become the de facto standard of the board. Open an MPSIMS thread, and we can conspire there on tactics and strategies for implementation.
I think you should do what Liberal said and just start using it yourself. It’s not like the SDMB is going make some formal rule or guideline as to what phonetic symbolism is going to get used so an ATMB thread is pretty pointless. Maybe post a link to an MPSIMS discussion on the subject in your last GQ language/pronunciation thread to get interested parties over.
You can enter any Unicode character by by using the form “&#XXX;” (without the quotes) , where XXX is the codepoint value in decimal. For example, here’s some IPA characters:
ʤ œ ɮ
And here is a table of IPA characters and their codepoint values.
Whether the characters will render depends on your browser and fonts, but most modern browsers on default settings should render them fine.
I dunno … I see that there are three symbols, but the first and last are default squares. The second one – a ligature of “o” and “e” – displays just fine.
I’m betting the miss rate on seeing all three of those symbols is close to 50%.
The IPA symbols in friedo’s link display just fine for me. I always wonder how far Unicode has been implemented in practical terms, given the technology in the hands of computer users right this minute. And how that may have changed over the past say 10 years. It would give a clearer idea of how necessary is it still to substitute an improvised workaround.