Protocol Action: 'Right-to-left scripts for IDNA' to Proposed Standard

[Slim Amamou]

On Sat, Feb 13, 2010 at 6:00 AM, Vint Cerf <vint at google.com> wrote:
> I am afraid that even UAX#9 will not solve the problem of random
> arbitrary text that is intended to be a domain name but which is not
> found in a context that allows reliable identification as such.
>
> vint
>

I agree here. It's an unsolvable problem anyway because communication
is context first, then information as a hint.

But what we are dealing with here is :
 - a domain name that looks different in different contexts where it's
easily identified as such
 - a loss of coherency in the structure of the domain name which leads
to confusion and inconsistency. For me it's not acceptable to display
L1.R2.R3.L4 as L1.R3.R2.L4 in an LTR context. I consider it  a defect.

-- 
Slim Amamou | سليم عمامو
http://alixsys.com