From: Patrick Andries (patrick.andries@xcential.com)
Date: Mon Feb 14 2005 - 07:59:48 CST
Michael Everson a écrit :
> At 01:12 -0500 2005-02-14, Patrick Andries wrote:
>
>> Are you considering such set of confusables for
>> Latin/Greek/Cyrillic/Tifinagh/Deseret/API/Old
>> Italic/Cherokee/Coptic/..., how large a set then? Maintaining such a
>> list of universal confusables seems like a lot of work, and by some
>> aspects very subjective.
>
>
> I agree, this is out of our scope, I think. It is "apologetic" and we
> already get attacked enough for this "security problem".
I agree for cross-script confusions : this does not seem like a real
security problem because, in practice, no one writes (or should write) a
word using multiples scripts (cases like the Cyrillic Q for Kurdish
should be fixed). In other cases, we must be careful not to duplicate
generic characters (like diacritical marks or symbols) across scripts.
>
> I mean, really. IDN is for, among others, the Chinese. Who is going to
> maintain the "list of confusables" for Extension B and Extension C1
> and all?
I also wonder. Isn't Unihan already a monstruous enough undertaking ?
P. A.
This archive was generated by hypermail 2.1.5 : Mon Feb 14 2005 - 08:01:25 CST