THanks for providing this information to John.
(FYI -- Just processing email from last month!).
jane huang wrote:
> Sorry for the late response to your earlier question about parallel writing
> system. Now I don't remember exactly what prompted your interests in
> parallel writing system. My guesses are that you want to know whether what
> Miyazawa discussed in her parallel writing system (Japanese, Chinese,
> Korean) presentation has any implications for ERC, or how ERC can be used to
> create metadata in those languages.
> Before talking about the metadata part, it may be helpful to clarify the
> difference between Japanese and Chinese. Japanese parallel writing system,
> Kanji and kana, is a native/inherent system. Both Kanji and Kana are
> integral parts of the writing system. If you open any Japanese book, you can
> easily spot them side by side. It's a necessity to supply kana when certain
> kanji is used. Because Kanji often have different meaning under different
> context. For instance, in a title, when certain kanji is used, Kana is used
> to denote the meaning of the kanji (word/phrase) in the context of the
> title. For personal name, product name, company name, etc., it is common to
> see both. It makes sense for Japanese to be concerned with the
> representation of this dual system in metadata.
> Different from Japanese, Chinese pinyin system (a Romanization system to
> denote the pronunciation of characters) is an add-on. It's a phonetic system
> used to Romanize characters, and often as a learning tool for people to
> learn Chinese. People who can read Chinese characters don't need it for
> comprehension purpose. But I noticed in North America, pinyin is always
> supplied in the catalog record of Chinese books. Semantically speaking, it
> is not a necessity.
> Besides this basic difference, both languages supply Romanization
> (kunrei-siki for Japanese and Pinyin for Chinese) in metadata and cataloging
> as regular practice. In the context of international information exchange,
> Romanization has its special functions. From the cross-walk or mapping
> perspective, this phenomena is also worth noting. I guess other non-roman
> languages (Arabic, Russian, etc.) may also have Romanization issues. If the
> discussion is in the context of Romanization, it would be different from the
> Japanese case. At the metadata implementation level, they may be treated as
> the same "parallel" or "alternative" or "multiple" values. As to how they
> can be specified in the abstract model, I have no idea how to do it.
> I don't know Korean, so can't comment.
> Now to ERC. Since ERC allows multiple values, parallel writing can be
> treated as "peer" value. Semantically it is slightly different from the
> English equivalents of "peer" values. But think of it, in the essence of
> "value", they do represent the same "who" or "what". That is what I can
> think of relevant to ERC. Let me know if you need to know more on this.