Re: [idn] Why follow IDNA with UTF-8?

[Martin Duerst]

At 14:05 01/07/15 -0700, Paul Hoffman / IMC wrote:

>[T]he editing interface to the host file does. That is pretty trivial: a 
>single script that converts the native-charset file (which does not have 
>to be UTF-8: it can be any native charset) to the host file. It took me 
>about 5 minutes to write mine in Perl (using an existing ACE<=>UTF-8 routine).

That's five minutes for each different format. And then that's a few seconds
every time a conversion is needed. Nobody in the US would want to do that
for US-ASCII. For US-ASCII, the widest variety of Internet protocols is
designed so that no such additional conversion step is necessary.
And for those protocols and formats where it's not the case, everybody
complains about, and for good reason. Why is it so difficult for somebody
like you to understand that the rest of the world would also benefit
greatly from this, and that it's a bad idea to deny them such benefits?


>>  > Have you now shifted
>>>  from "ACE then UTF-8" to "never ACE"?

I for one always claimed that ACE is a big step backwards.


Regards,   Martin.