Re: Standardizing on IDNA 2003 in the URL Standard

On Thu, Jan 16, 2014 at 4:32 PM, Mark Davis ☕ <> wrote:
>> Or A to a?
> I'm assuming that you mean the ascii characters (I'm not going to check
> whether you have just look-alikes.). ASCII case mapping is covered at a
> different level.

It is covered by the UTS46 mapping table actually.

> I don't think clients would stop
> mapping, and IDNA2008 permits it. That's why I said "
> plus a mapping layer that can be used with straight IDNA2008"

Okay. It is sometimes a bit unclear to me if people are considering
the whole of UTS46 to be transitional or just its Transitional

>> Another wish I have is that the algorithms are a bit clearer in terms
>> of input and output. What argument does ToASCII take? What about
>> ToUnicode?
>> E.g. how would you replace "domain to ASCII" and "domain to Unicode"
>> in with UTS46 and
>> ensure the algorithm still has the same kind of expected output?

Right, that algorithm does not really define what input it takes, that
it takes UseSTD3ASCIIRules as flag, that it takes some kind of marker
to indicate the processing desired, and that it returns something.

> If there are specific areas where you find the spec unclear, I suggest that
> you provide feedback as instructed at the top of the spec. Subsequent
> versions can then clarify those points.

I just did. And I think I might have before as I remember seeing that
form. Might I suggest getting a public list or Bugzilla instance so
both sides can track what is going on?


Received on Friday, 17 January 2014 14:04:13 UTC