Re: LOTEP (was: Re: A single font can display ANY alphabet, pictograph, or rune)

--- René Uittenbogaard <ruittenb@...> wrote:

<snip>

> 1. How does the LOTEP system handle dots? Do they
> count as a single
> endpoint? What would the LOTEP number for
> (helvetica) |i| be - 00302 ?
>
I don't recall what I originally did. I'm inclided to
count a dot as a separate piece but not as an
endpoint.

> 2. Your second example on the page (the 22631 glyph)
> seems to have an
> "angle" (discontinuity in the first derivative, if
> it were a curve). I
> understand that angles are not counted as "Even
> crossings". Would it be
> an idea to add an Angle count, defined as the number
> of points where
> there is a discontinuity in the first derivative
> along a line (otherwise
> just a 2-crossing - I mean that Odd and Even
> crossings don't count as
> Angles)
>
I considered this when I first developed the system,
but decided against it simply because in many cases
the angle is actually optional. The upper case "A" may
be written by hand with a rounded top rather than an
angular one and it is still counted as "A". Granted
the only difference between "U" and "V" is the angle,
but most letters are not distinguished by the absence
or existence of an angle. Even "E" can be written as a
rounded "C" with a horizontal bar and still remain
recognizable as the same letter as "E" written with
angular corners.

As long as something as variable as whether a given
corner is to be counted as angular or not is included
in the system then that opens the door for a greater
rate of variability in how the characters are encoded,
making it more likely that a given user would encode a
given glyph in a different way from another user.

--gary


> René
>