[UA-discuss] ODG: Re: UA issue

Andrew Sullivan ajs at anvilwalrusden.com
Fri Jan 22 00:57:36 UTC 2016


On Fri, Jan 22, 2016 at 12:45:46AM +0000, Mark Svancarek wrote:
> Good point.  But I don't think that "proper" captures the fact that the string is in use and could be expected to return a result...
> 

Maybe I'm not understanding, then, but is the idea that this is an
all-ok-in-protocol string (I'm being coy, because IDNA is defined for
labels, for instance) or is it not necessarily that?

It's worth noting that IDNA2008's idea of IDN is clear: it has only
A-labels, U-labels, or NR-LDH labels.  Since every A-label is a
U-label and conversely, this boils down to "traditional labels, or
IDNA stuff".  The _problem_ with that is that some people use "IDN" to
mean things outside of IDNA2008, including IDNA2003-compatible names
and Unicode-on-the-wire names.  The latter two may contain labels that
are not U-labels in IDNA2008.

So, at least in the case of domain names, I'm trying to understand
whether we're talking about subsets of IDNA2008 rules that are
actually in use, or things that are actually in use that may not
conform to IDNA2008.

Thanks,

A

> -----Original Message-----
> From: ua-discuss-bounces at icann.org [mailto:ua-discuss-bounces at icann.org] On Behalf Of Andrew Sullivan
> Sent: Thursday, January 21, 2016 2:18 PM
> To: ua-discuss at icann.org
> Subject: Re: [UA-discuss] ODG: Re: UA issue
> 
> On Thu, Jan 21, 2016 at 07:35:38PM +0000, Mark Svancarek wrote:
> > We have been discussing terminology and taxonomy, this is a good example.
> > 
> > I like “well-formed” and “syntactically correct” and “RFC-compliant” to describe strings which we expect UA-Ready applications and services to consume in RFC-compliant ways.
> > 
> > I think “valid” is a good way to describe strings which are not only syntactically correct but also in use in the ecosystem.
> 
> IDNA has the notion of PVALID (or PROTOCOL-VALID) for code points that are definitely allowed, so "valid" might sow confusion.  Perhaps "proper"?
> 
> A
> 
> --
> Andrew Sullivan
> ajs at anvilwalrusden.com

-- 
Andrew Sullivan
ajs at anvilwalrusden.com


More information about the UA-discuss mailing list