Punycode & IMA/EAI

Shawn Steele Shawn.Steele at microsoft.com
Thu May 22 01:43:16 CEST 2008

The non-IDN aware app is fine for hrefs, but the non-IDN aware app isn't going to handle bücher.de, even though it points to buecher.de, which a german speaker could nearly figure out for themselves.  Punycode doesn't help in this case.

Its nice to think that IDN is "backwards compatible" for non-aware apps, but in practice it is NOT backwards compatible once a human gets involved, humans can't convert Unicode to Punycode and they can't read Punycode.

- Shawn

-----Original Message-----
From: james.seng at gmail.com [mailto:james.seng at gmail.com] On Behalf Of James Seng
Sent: Wednesday, May 21, 2008 4:05 PM
To: Shawn Steele
Cc: Tina Dam; Martin Duerst; John C Klensin; Felix Sasaki; idna-update at alvestrand.no
Subject: Re: Punycode & IMA/EAI

I think (b) is most likely something we put in the specification as an
"error checking" so we could warn the users but would not stop it from
moving forward.

afterall, a non-IDN awared app will behave as (a) .. and it is for
that specific behavior, ie backward compatibility that we have
IDNA/punycode solution today.

-James Seng

On Thu, May 22, 2008 at 4:06 AM, Shawn Steele
<Shawn.Steele at microsoft.com> wrote:
> We've digressed from the point that I'm trying to make, which is that requiring lookup of unknown punycode names is not a solution for forward compatibility, unless you also have a successful way of getting that punycode from the actual Unicode.  The cases where punycode will solve problems isn't going to touch the usability, and as a sales rep or whatever I cannot use a Unicode name if it isn't usable to the people reading my card.
> So my opinion is that either:
> a) there shouldn't be a requirement to lookup unknown punycode to solve forward compatibility issues because it only helps a subset of the problem, or:
> b) if there is such a requirement to lookup unknown punycode, then there must be a provision to also allow converting Unicode code points to punycode.  This isn't a problem if the Unicode is "normalized".  It could be a problem if the Unicode is in a case mapped form or other form that might not make the conversion straightforward.
> - Shawn
> -----Original Message-----
> From: Tina Dam [mailto:tina.dam at icann.org]
> Sent: Wednesday, May 21, 2008 12:57 PM
> To: Martin Duerst; John C Klensin; Felix Sasaki; Shawn Steele
> Cc: idna-update at alvestrand.no
> Subject: RE: Punycode & IMA/EAI
> I don't think I understand why this email-list/group is spending time discussing how popular or not popular IDNs are going to be....the matter of the fact is that IDNs exists at the second level under numerous TLDs, and there are processes underway aiming at making it available at the top level as well.
> So, there really is no need to talk about "whether or not" or "how successful or useful" - the development on both technical and policy side are underway and as far as I believe that decision was made some time ago and is not up for debate today.
> If you want to talk about usability or try things out, one good place to do that is on the IDN wiki: http://idn.icann.org
> Tina
> _______________________________________________
> Idna-update mailing list
> Idna-update at alvestrand.no
> http://www.alvestrand.no/mailman/listinfo/idna-update

More information about the Idna-update mailing list