Re: Unused code positions and mapping to Unicode

From: Asmus Freytag (
Date: Fri Aug 06 1999 - 20:40:39 EDT

At 03:36 PM 8/6/99 -0700, Geoffrey Waigh wrote:
>But if you don't know how to correctly map it, putting it somewhere else
>is bad too. People can kludge it with remapping out of the incorrect

If you use distinct PUA (=aka EUDC) positions for all holes in all legacy
sets, you can in principle round-trip anything, and kludge to your hearts
content (and with the new 128K user characters in surrograte space, you can
really party.

>spot afterwards, though if they have multiple source codepages that
>filled in their holes differently, that might not be such an easy task.
>At least with FFFD, the user knows they have to upgrade their software -
>which is another reason that having Unicode properties and libraries
>in a system supplied standard package rather than done in each
>application is a Good Thing. Alas, the politics on that one will
>dog us for many years.

I see that you are a supporter of the Principled position (P).


This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:50 EDT