Re: Combining Box Characters?

From: verdy_p (
Date: Mon Jan 04 2010 - 00:17:42 CST

  • Next message: Michael Everson: "Re: Latin-script keyboard layout (was RE: Quick Question About Korean Input Methods)"

    > De : "Doug Ewell"
    > Leslie Turriff wrote:
    > > Certainly, but that does not involve combining characters.
    > I missed the point that by "combine" you meant you wanted to overlay
    > certain box-drawing characters with others. AFAIK there is no standard
    > Unicode way to do this. Box-drawing characters are not combining
    > characters, and backspacing to display one character on top of another
    > is not something you would normally do in Unicode.
    > If you want to draw nice diagrams, you can only go so far before the
    > time comes to abandon the notion of "Unicode art" (cf. "ASCII art") and
    > use a real 2- or 3-dimensional diagram-drawing tool.

    May be there's confusion between the "Combine" key on the keyboard, which is used as part of an input method (which
    could use other GUI designs as well, or other tools like gestures on a tablet with a stick, or with the mouse, or on
    a touchscreen or a touchpad).

    But keyboards and input tools are definitely not the same species as data encoding. We don't even print text the way
    we draw them on a paper, and even when we change the inked tool (pens, brushes...), we use different technics and
    change the order of drawing for various practical reasons.

    The needs are completely orthogonal. For this reason, keyboards don't have necessarily to follow the logical
    encoding order. What they have to do is to provide a convenient and fast way to enter text with fingers. Ergonomy
    drives the design, not the efficiency of encoding for computer programs.

    Computers don't have hands and fingers! but they can access to many more keys without errors, and in a much faster
    way: it's up to the comptuer to adapt to our ergonomy, even if the data will be encoded another way for more
    convenient massive data handlign by computer programs...

    But even still, computers have lots of difficulties to represent and correctly manage what we can do easily after
    some basic training, because they are unable to understand (i.e. generalizing and associating concepts by analogy,
    then using this extra knowledge founded by "experience" into creative/generative ways) : in general we demand
    exactly the opposite to computer programs: we want them to remain restricted within bounded domains and to apply the
    rules strictly, so that they can be reproduced identically.

    This clearly justifies that data input methods (for humans and according to ergonomy) remain orthogonal to data
    encoding and handlign processes: there can safely exist a conversion from one representation to the other, and this
    is the role of the input drivers (for e.g. keyboards, mice, and other tools).


    This archive was generated by hypermail 2.1.5 : Mon Jan 04 2010 - 00:19:06 CST