On 11/8/2012 4:53 PM, Murray Sargent wrote:
Mark E. Shoulson <m...@kli.org> wrote: Mirroring tends to be done for glyphs
that are used in *pairs*,
open/close things and such.
Not invariably; consider the integral and summation. They don't have mirrored
counterparts and many other mathematical symbols don't either.
Mirroring is reserved for symbols that are unambiguously used in
opposite direction. The case was made that arrows might not qualify
because it wasn't clear whether they were denoting an inline direction
or a real-world direction. That decision must stand, for reasons of
document portability and compatibility over time, but as time passes,
I'm less comfortable that it was the best way to treat arrows. Those
second thoughts aside, anytime you have a non bilaterally symmetric
image used as a symbol it's not clear that mirroring that image would be
either required or acceptable.
In mathematical notation, the accepted practice is to mirror, that's why
mathematical symbols get treated the way they do.
A./
