Re: [VCARDDAV] Review of draft-fukuda-vcarddav-phonetic-transcription-01.txt

[Gren Elliot <[e-mail]>]

Hi Fukada-san and DataPacRat,

> I'm not sure if you meant to use the 'Type' parameter in your second
> example; instead of
>
> PHONETIC-FULL-NAME;TYPE=IPA-X-SAMPA:jamada ta4oM;;
>
> you may have meant something more like
>
> PHONETIC-FULL-NAME;LANGUAGE=fonxsamp:jamada ta4oM;;
>
> or
>
> PHONETIC-FULL-NAME;LANGUAGE=jp-fonxsamp:jamada ta4oM;;

I think the original example is more accurate.  "ja" is an appropriate
setting for a language but I don't believe either "fonxsamp" or
"jp-fonxsamp" fit the description of a language.

I'd like to look at 2 examples to illustrate another issue with
treating LANGUAGE as the most important parameter.

If you understand Japanese and specifically Hiragana/Katakana work,
you can guess that the text of the following will be only Hiragana or
Katakana.  However, nothing actually says that, and I think that is
crucial missing information.

PHONETIC-FULL-NAME;LANGUAGE=jp:<U+3084><U+307E><U+3060>
       <U+305F><U+308D><U+3046>;<U+3084><U+307E><U+3060>;
       <U+305F><U+308D><U+3046>
      ; <U+XXXX> denotes a UTF8-encoded Unicode character.

However, even if you understand English:

PHONETIC-FULL-NAME;LANGUAGE=en:Siobhan St. John Smith

wouldn't help you pronounce that name.  Most English people would
probably guess wrong.

I think TYPE or MEDIA-TYPE are more appropriate.  Given Cyrus Daboo's
familiarity with the RFC process, I suspect his suggestions are likely
to be the better of these two.  I propose MEDIATYPE in preference to
TYPE.  The default should be:

MEDIATYPE=text/ipa; option=x-sampa

The full Hiragana example would then become:

PHONETIC-FULL-NAME;MEDIATYPE=text/hiragana:<U+3084><U+307E><U+3060>
       <U+305F><U+308D><U+3046>;<U+3084><U+307E><U+3060>;
       <U+305F><U+308D><U+3046>
      ; <U+XXXX> denotes a UTF8-encoded Unicode character.

Thanks and regards,
Gren