This spoofing and security thread

From: Michael Everson (everson@evertype.com)
Date: Mon Feb 11 2002 - 05:22:51 EST


Bernard Miller wrote:

> Instead of proclaiming Unicode is not to blame

Unicode is not to blame. The facts of the history of the development
of writing systems the world over yields a result that you are
uncomfortable with.

Unicode exists to encode the world's writing systems

> > it would be a good idea for Unicode to provide some character
>properties that could help an implementation reduce spoofing.

Like what? A set of rules that says "one of these things is not like
the others"? How are you going to tell which letters belong? Visual
comparison of what is in the charts? Those images aren't normative.

Myanmar letter WA! Digit 0! Wait! Look! There's a Canadian Syllabics
letter that looks just like the letter L! What other superficial
glyph similarities can we find? What happens when we change fonts?

This thread is a colossal waste of time. If somebody wants to type my
surname with Cyrillic characters in it there is nothing I can do
about it, and trying to prevent it is not only impossible but it's
nonsensical. If there is a problem, it isn't Unicode's. Unicode is a
big huge typewriter. We can't prevent people from typing with it what
they wish.

-- 
Michael Everson *** Everson Typography *** http://www.evertype.com



This archive was generated by hypermail 2.1.2 : Mon Feb 11 2002 - 05:05:54 EST