Re: Proposal to encode an EXTERNAL LINK symbol in the BMP

From: Jukka K. Korpela (jkorpela@cs.tut.fi)
Date: Mon Jul 24 2006 - 06:17:38 CDT

  • Next message: Jukka K. Korpela: "Re: Proposal to encode an EXTERNAL LINK symbol in the BMP"

    On Mon, 24 Jul 2006, Michael Everson wrote:

    > It seems to me that the character as proposed is a symbol found as a text
    > element in a wide variety of printed and on-line sources.

    I can't remember having anything like the proposed EXTERNAL LINK symbol
    appear in printed sources, except in printed copies of web pages that use
    an image resembling it. In online sources, as I mentioned, there is
    considerable variation in the images used for the purpose. I don't think
    we have sufficient grounds for deciding that a particular _character_
    can be recognized.

    If a character were defined, it would at most be used as _an_ external
    link marker, rather than as _the_ marker. Despite the reservations I made
    in my previous comment, there would be sufficient grounds for defining a
    character _if_ a consistent usage and shape could be detected. However,
    as I see it, this is a matter of varying usage of icon-like images as
    external link markers, and defining a character shaped to match one of
    the designs would not sound good. It would mean a policy decision to
    favor one of the designs, and the use of such markers in general, rather
    than a neutral decision to encode a symbol that has actual widespread
    usage in texts.

    Besides, Unicode is about plain text, and links are a hypertext matter.
    Is there some need to use an EXTERNAL LINK symbol in _plain text_?
    Whenever you have links, you are working with hypertext, not plain text.
    The mere possibility of printing a hypertext document or saving it as
    flattened to plain text does not sound like a sufficient reason for
    encoding EXTERNAL LINK as a character. If an image is used as an external
    link marker, it can be printed too, and in formats like HTML, it can carry
    an "alt" attribute that specifies the textual replacement.

    -- 
    Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
    


    This archive was generated by hypermail 2.1.5 : Mon Jul 24 2006 - 06:25:11 CDT