Re: [idn] Debunking the ACE myth

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

>            2002       2003       2004       2005       2006
>           +----------+----------+----------+----------+----------+
>    ACE   A|**********|**********|********* |********  |******    |
>          B|          |          |          |          |          |
>          C|XXXXXXX   |XXXXXXXXXX|XXXXXXXX  |XXXXXX    |XXXX      |
>           +----------+----------+----------+----------+----------+
>    UTF-8 A|******    |***       |*         |          |          |
>          B|XXX       |XX        |          |          |          |
>          C|XXXX      |XXX       |X         |          |          |
>           +----------+----------+----------+----------+----------+
> 
> The X's represent mail reply failures: row B for built-in replies, row C
> for copy-and-paste. The *'s in row A represent annoying IDN displays.
> All the numbers are guesses; we need further analysis.
> 
> The ACE myth is that there are no X's in the ACE boxes. What's true is
> that there are no X's in row B in the ACE boxes.
> 

thanks, that's a useful illustration.

I find it interesting that it assumes that MUAs will be upgraded more
quickly with UTF-8 than with ACE.  I would assume just the opposite - 
that UTF-8 is more immediately disruptive because the failures that 
it does cause are more numerous, and thus a larger barrier to adoption. 

Also, the X's in row C look larger than the spaces in row B - but 
built-in replies are surely far more common than copy-and-paste replies.

The graph also does not consider other kinds of failures due to UTF-8 -
e.g. failures of mail (including nondelivery reports) to be delivered
because one of the intermediate systems either couldn't deal with
the UTF-8 or because it munged the UTF-8 to the point that subsequent
systems couldn't make sense of it.


Precisely because of the copy-and-paste problem, I expect that
vendors will quickly upgrade MUAs to accept native IDN input 
regardless of whether ACE is used as the on-the-wire encoding.
Whether a large percentage of their customers upgrade quickly is
a different question-  but even if a few percent require that 
functionality, that's sufficient motiviation to fix the product.

Again, I don't think this is a either/or choice for this group-
most of us are convinced that we are going to need ACE for some
applications (at least during a transition period), and that we 
are going to use UTF-8 on the wire within other applications.  
Those choices will be made by other working groups.  

Keith