Re: A conlang idea rolling around in my head

--- Greg Johnston <greg.johnstons@...> wrote:
> Yes, that was my second question. As many of my
> glyphs would start with a
> 0 or 1 as a LOTEP first digit, maybe there could be
> another digit at the
> beginning for the number of seperate lines.
>
> i.e. A has three. N also has three. # has four. This
> would be the number
> of constituent line segments (including curves)
> which would be seperated
> by joints in the glyph.
That might work as long as there are not too many
ambiguous cases where a glyph is drawn with something
that could be considered a "joint" or might also be
drawn as a rounded corner.  But as long as care was
taken to avoid ambiguity then something like that
would distinguish between "U" and "V" which would
otherwise have the same index number.

Another thought I had was to draw an imaginary line
through the glyph from top to bottom (or left to
right) and count how many glyph lines it cuts through.
 "E" and "F" have the same lotep number, but an
imaginary vertical line through "E" cuts 3 glyph lines
while the same vertical cut through "F" only cuts 2
glyph lines.  So "F" would have a "vertical cut
number" of 2 and "E" would have a "vertical cut
number" of 3.

Horizontal cut number could also be used.

Another thought would be to distinguish the number of
endpoints on the top, left, right and bottom edges.
"E" has all its endpoints facing the right edge while
"Y", even though it has the sme lotep number, has 2
endpoints on top edge and 1 on the bottom edge.  This
would give a different index number to the mirror
image of a glyph which the original lotep number does
not.

Any of these could be appended to the original 5 digit
lotep number to give finer resolution to the index
numbers.

Any other brainstorms anyone?