Re: [idn] UTF-8 / RACE

[Keith Moore <moore@cs.utk.edu>]

> >ACE affords incremental deployment much better than no-ACE.  Suppose I
> >am considering getting an IDN for my domain.  With ACE, this will make
> >things better for some users (who have upgraded their clients to decode
> >the ACE) and worse for others (who have old clients and will see ugly
> >ACEs) but nothing will actually break (mail will get through, web pages
> >will load, etc).
> 
> This is completely wrong because it disregards the human face of the
> Internet. 

I don't see how you get that idea.  Both ACE and UTF-8 solutions would
accept and display IDNs in native form for upgraded applciations.
Human users of upgraded applications would see no difference in the
input or display of IDNs.  

The only difference that they would see is that:

- the ACE solution gets deployed more quickly
- the ACE solution is more reliable in the near term
- the ACE solution works better with interoperating with non-upgraded clients

you are correct that the Internet is a medium for human communications.
but humans need for those communications to be reliable, and going directly
to UTF-8 disrupts that reliability.

here is one way of comparing the two:

- if we use ACE, human users will get annoyed when those ACE names are
  displayed by non-upgraded tools, and when they leak
  (as for example when an email header address is copied into the message 
  text)

- if we use UTF-8, human users will get annoyed when those UTF-8 names
  are displayed by tools that don't support UTF-8, and when they
  leak to environments that don't display UTF-8, and when they do not
  work as reliably as before (because the UTF-8 is causing sutble
  bugs in software that was written to expect ASCII), and/or they cannot
  interoperate with tools that are not upgraded.

people will complain no matter which path we take.  but the former approach
works better and produces fewer problems.

Keith