Re: Localizable sentences (from Re: On the possibility of encoding webdings in Unicode)

From: Neil Harris (
Date: Fri Jan 28 2011 - 05:03:50 CST

  • Next message: William_J_G Overington: "Re: Localizable sentences (from Re: On the possibility of encoding webdings in Unicode)"

    On 28/01/11 10:10, William_J_G Overington wrote:
    > On Thursday 27 January 2011, Doug Ewell<> wrote:
    >> I do not believe localizable sentences will ever be appropriate for standardization in a plain-text character encoding, regardless of the size of the community of users. Perhaps they would be appropriate for standardization somewhere, but not in a character encoding.
    > If localizable sentences become encoded in Unicode plain-text character encoding then they could be used intermixed with ordinary language.
    > Suppose please that the following localizable sentence were encoded using one codepoint of plane 7.
    > U+7XXXX Where can I buy a meal without any gluten in it please?
    > For example, please consider that an English-speaking person in an Italian town wants to ask "Where can I buy a meal without any gluten in it please?" then he or she could use the U+7XXXX codepoint and throw it from an iPad to an iPad of an Italian-speaking person who could reply with the name of a restaurant and its address.
    > If localizable sentences for giving movement directions were also encoded in Unicode then the Italian-speaking person could throw back directions that he or she had composed from an iPad menu in Italian and those directions would be displayed in English on the receiving iPad.
    > William Overington
    > 28 January 2011


    I can trivially enumerate tens of millions of such simple sentences, and
    there are substantially less than a million free code points left.

    Since it would seem rather inconvenient to have no way to encode new
    characters into Unicode once the code point space has been filled with
    sentences, could you perhaps create a list of the sentences you believe
    most important to encode, and a rationale as to why these particular
    sentences, and not others, are worth encoding?

    Also, given that memory is cheap now, would it not be a simpler way to
    achieve your goal just be to create a list of such sentences in
    English,together with a list of their translations in each language, and
    distribute it widely? A file containing a million such sentences, each
    translated into a hundred languages, would compress to a file size of
    roughly a hundred megabytes, which could trivially be stored on any
    modern hand-held device, and would eliminate any need to make encoding
    decisions: you would simply send the English-language version as a key.

    -- Neil

    This archive was generated by hypermail 2.1.5 : Fri Jan 28 2011 - 05:07:02 CST