Again you have fallen into the common underestimation of the upgrade work
required for ACE to work perfectly (including display). What I see is
that
applications are already having some trouble going from ASCII and local
encoding to UTF. ACE in essence requires the conversion from local
encoding
toUTF first and then toACE. I see trouble here...
Yeah, but I suspect it's smaller than you think; once you've got UTF-8 in
place, adding the UTF-8/ACE conversion would be a lot simpler. In any
case, I don't think we can do without ACE, at least as a transition, which
means that a later transition from ACE to UTF-8 wouldn't reduce the
complexity much (you'd still need to talk to ACE-based servers).
That's all the points there, once we've got UTF8 in place, so if we have
UTF8 in place then we only need the ACE as a transition, then we can slowly
phase out ACE later, and be able to use UTF8.
That's what me and Edmon has been talking about all the time, we need ACE as
transition and move to UTF8 as smooth as we can. So an internet draft for
this transition is needed, we will be posting a draft about this soon.