November 1997 To users of WinCALIS/UniEdit Japanese: WinCALIS/WinCALIS Author beginning with version 2.3 and UniEdit beginning with version 1.3, handle Japanese text slightly differently. Now all Japanese text is processed and stored as 100% Unicode text. Previously, in order to assure that the glyphs (character pictures) in our Unified Han font always were culturally correct for the Japanese context, about 300 Japanese kanji characters were encoded with Private Use Area codes. (These kanji characters can be viewed by going to the block F000 - F1FF in the Unicode browser (Unicode Input under the Edit menu.) This was necessitated because the Unicode Han character set in some cases treats differences in character shape as character distinctions, with separate code points, but in other cases makes no distinction, treating them as minor glyphic variations. So, for example, Unicode makes a three-way distinction between the three different glyphs for Kangxi Radical 63, meaning door, U+6236, U+6237, and U+6238 (Japanese 'to'), but usually makes no distinction between the characters using this radical. Thus since there is only one code point for the character U+6240 meaning "place," in the past we used a distinct private code point F013, for Japanese 'tokoro' (and Traditional Chinese 'suo3'), using U+6240 only for Modern Standard Chinese orthography, as used in the PRC and Singapore. Now, however, as the rest of the world has increasingly joined us in using Unicode as its character coding standard, and our users are increasingly exchanging Unicode text with other Unicode applications (e.g., Microsoft Office 97 applications, Netscape, etc.), it is important that we also use standard Unicode throughout the Japanese Han character set. We are now distributing separate "Japanese-glyph" versions of two of our CJK character font files, WC2CJK1J.FON and WC2CJK2J.FON. These should be installed in your copy of Windows. See the separate README.TXT file in the temporary directory created when you extract files from the archive WC224FNT.EXE. In order to make use of these new font files to assure that the now 100% pure Unicode characters in your Japanese text continue to appear culturally correct in the Japanese context, you should be sure that your main default character-glyph table WCFNT.INF is the correct one for Japanese. To confirm that this is the case, after upgrading to the latest version of WinCALIS/UniEdit,you can open the file WCFNT.INF (in the main WinCALIS program directory) in a plain text editor such as the Windows Notepad, and be sure that the font "Mincho" is specified for a whole block of font files beginning around line 75. If it is not, you should copy the file WCFNTJ.INF to WCFNT.INF, overwriting the file currently with that name. In order to restore the Modern Standard Chinese shapes of these 300 or so characters, you can overwrite WCFNT.INF with WCFNTC.INF, which is a copy of the original WCFNT.INF. Any existing files using Private Use Area codes may be converted using either the "Pure Unicode" code filter in the File, Open and File, Save dialogs, or the DOS batch conversion utility WC2&16BT.EXE, with the code filter WC&PURUN.TAB. See the topic "Conversions" in the documentation for details. WinCALIS Authors using Japanese can assure that their scripts will always use the culturally correct Japanese fonts by adding the line L=Mincho at the beginning of each .CAL file. This can be done in WinCALIS Author by inserting the line in the Pass Through window of Task 1. We regret the inconvenience this will cause some users. We would be pleased to assist in the conversion process. If you send us your files, we can convert them for you and return them to you. japanese.txt rev. 11/1/97