Re: [Maybe OT] localized names of the Unicode Control characters

From: Philippe Verdy (
Date: Tue May 27 2003 - 08:01:08 EDT

  • Next message: Karl Pentzlin: "Re: Dutch IJ, again"

    From: "Marco Cimarosti" <>
    > ISO 10646 has the French translation of all the character names. In most
    > cases, the French names are just literal translations of the English ones
    > but, in a few cases, they have a completely different wording, which can
    > provide a better starting point for translations in other languages.
    > As you say, those names are often meaningless labels intended for techie
    > people. In a user interface, you should probably invent more friendly terms
    > (e.g. "joiner", "splitter"), leaving the literal translation (or even just
    > the untranslated English term) only in the manual or help file.

    If you look in the French localization of the Windows XP version of the "charmap" tool, you'll see that Microsoft displays these French translations for character names. There are however some "strange" translations that lack a common formal format that allows easier searching for related characters.

    But this demonstrates that the normative names are only useful when discussing the standard implementation, but not relevant for any actual implementations that must fit user expectations.

    I did not know that ISO10646 lists French versions of these names, andwonder if this is normative. If so, why aren't these French names listed by some derived Unicode file (which would combine the UCD with the file ISO10646 French names) ? There are cases where the use of a single language is not enough to clearly indicate the semantic of a character with a short name, and a translation may sometimes be useful as an additional reference to avoid ambiguities, or simply to match a common typographic convention widely used by publishers to designate these characters or symbols.

    This idea goes for beond a simple translation, but it's a true localization issue, where the language is not the only relevant information, but also the domain of application (which may have some strong historical conventions when designating some characters). For now some aliases are listed only in the detailed charmaps and only the English ones, but they don't appear in the core or derived files of the UCD.

    Is there some work in progress (probably with ISO10646 cooperation) to include files for contextual character name aliases in some new UCD files ?

    This archive was generated by hypermail 2.1.5 : Tue May 27 2003 - 09:49:19 EDT