Re: Medievalist ligature character in the PUA

From: William_J_G Overington (
Date: Mon Dec 14 2009 - 04:13:32 CST

  • Next message: John H. Jenkins: "Re: Medievalist ligature character in the PUA"

    On Monday 14 December 2009, André Szabolcs Szelp <> wrote:

    > If they are not used for interchange
    > (wikisource sending you data to display on your screen is
    > interchange already!), how would they faciliate correct
    > display, as compared to the status quo?

    Yes, I have not quite got it right yet.

    I was meaning not used for interchange in such things as producing documents for publication. I was thinking of the use of the precomposed ligature code points for producing a display on the screen as local processing, in that the version with the ZWJ encoding would be used when storing the document and the precomposed ligature version only generated when needed at display time. Yet I accept that I have not yet managed to express what I am meaning in a precise manner.

    I am trying to find a middle ground between the Unicode Consortium not wishing to encode any more precomposed ligatures except in special circumstances and what appears to be a need to have some standardized codepoints for ligature glyphs within fonts without using a Private Use Area encoding. It seems to me that there is a lot of unused space in the Unicode code point map and that if there is a need by some people for some of those code points to be used for precomposed ligatures then perhaps some progress can be made. I appreciate that there could be side-effects such as decomposition tables and so on, so I was thinking that maybe the problems could be avoided by the precomposed ligatures not being decomposable. I am not an expert in Unicode yet I am hoping that maybe experts can identify any side-effects that could be produced and perhaps also devise ways around them so that some more precomposed ligature glyphs could become mapped into regular
     Unicode yet that process would not cause problems.
    > Please note, but I'm sure you frequenting font design
    > forums know, that in modern fonts glyphs don't need to
    > have actual unicode (or any other [external] encoding)
    > codepoint assigned.

    Well, I know that in OpenType fonts that a ligature glyph does not need to have a Unicode codepoint assigned to it, yet I also know that it may have a Unicode codepoint assigned to it if the font designer so chooses.

    I am aware that some font designers do not like to map ligature glyphs and alternate glyphs in OpenType fonts into the Private Use Area and that some font designers do like to map ligature glyphs and alternate glyphs in OpenType fonts into the Private Use Area.

    Mapping a ligature glyph or an alternate glyph of an OpenType font into the Private Use Area can be useful as it allows someone who is using budget software packages to access the glyphs and produce hardcopy print outs and graphic files using one or more of the glyphs.

    It does seem to me that something could be gained and nothing lost if some precomposed ligature glyphs were to become encoded in plane 14. I accept that I am not expert on the side-effect implications of such an encoding and hopefully advice will emerge in this thread.

    William Overington

    14 December 2009

    This archive was generated by hypermail 2.1.5 : Mon Dec 14 2009 - 04:15:27 CST