zenkaku, hankaku

[Peter Clark]

Dear Yakkers,
zenkaku, hankaku
Translators are constantly faced with these two words, and a search in the archive for "half-width" turned up some ancient conversation with eminent members.
 
In this particular case, it is in a discussion of what happens to a certain set of Unicode characters when they are output. Jim Breen in April 2004:
Note that double-byte/single-byte for 全角/半角 O N L Y applies to
Shift_JIS coding. 半角 characters take 2 bytes in EUC-JP, 3 bytes in
UTF8 and up to 7 bytes in ISO-2022-JP.
I'd only translate them as half-width and full-width.

So as the number of bytes is almost never single or "double", and the important information is the appearance of the characters on screen or in print, then any mention of bytes when translating theses terms should be limited to discussions of Shift-JIS, and in all other cases should be translated as half-width and full-width.
 
In the same discussion, Edward Lipsett suggests using normal width and double width, as from the English perspective, this is what is happening. But as this distinction is almost always with reference to Japanese, and is directed at an audience with experience using Japanese, I disagree, and would prefer to keep with half and full.
 
Have there been any shifts in this area over the last decade or so? Has Unicode made bigger inroads into the Japanese input/encoding market?
 
Regards,
 
Peter Clark

[Herman]

The Unicode block containing 半角カナ　and 全角ローマ字 is called
'Halfwidth and fullwidth forms', so there is no change, at least as far
as the terminology goes.

Herman Kahn

[Peter Clark]

Thanks Herman. I notice you use the terms as one word, unhyphenated.
That is probably a shift, albeit an expected one.
 
Regards,
 
Peter Clark

>
> The Unicode block containing 半角カナ　and 全角ローマ字 is called
> 'Halfwidth and fullwidth forms', so there is no change, at least as far
> as the terminology goes.
>
> Herman Kahn
>

> --
> --
> You received this message because you are subscribed to the Honyaku Mailing list.
> To unsubscribe from this group, send email to honyaku+u...@googlegroups.com
> For more options, visit this group at http://groups.google.com/group/honyaku?hl=en?hl=en
>
>
>

[Brian Chandler]

Peter Clark wrote:
> In [2004] Edward Lipsett suggests using normal width and double width, as from the English

> perspective, this is what is happening. But as this distinction is almost always with reference to
> Japanese, and is directed at an audience with experience using Japanese, I disagree, and would
> prefer to keep with half and full.

(I'm being idle and not looking anything up, so fwiw)

"Full-width" is clearly inappropriate as an English description -- it
should have been "full-box" (or "full-kaku" or something). But too
late: the Unicode standard describes all this stuff as EAST ASIAN FULL
WIDTH this that and the other, so you have to make sure it is clear
from the context that we are talking about East Asian typesetting,
essentially. (Of course there is no difference between the two sets of
the Roman alphabet required for East Asia except a typographical one,
but Unicode is engineering not philosophy, and it is obliged to
provide round-trip compatibility with legacy character sets.)

I do not believe that the real Roman alphabet (i.e. "ASCII") is ever
called "half-width", since it isn't; so you might sometimes have to
translate 半角 as "half-width East Asian characters and/including the
Latin alphabet", or somesuch.

> Have there been any shifts in this area over the last decade or so? Has Unicode made
> bigger inroads into the Japanese input/encoding market? Regards, Peter Clark

Yes. So I think you are definitely right to discard the "bytes" stuff.

Incidentally, I certainly would not follow the Unicode people on
matters of English style, such as hyphens, or you will end up in block
capitals.

Brian Chandler