- From: Andrew Sullivan <ajs@anvilwalrusden.com>
- Date: Fri, 23 Aug 2013 11:23:10 -0400
- To: Mark Davis ? <mark@macchiato.com>
- Cc: Vint Cerf <vint@google.com>, Gervase Markham <gerv@mozilla.org>, Peter Saint-Andre <stpeter@stpeter.im>, "uri@w3.org" <uri@w3.org>, "PUBLIC-IRI@W3.ORG" <public-iri@w3.org>, Shawn Steele <Shawn.Steele@microsoft.com>, IDNA update work <idna-update@alvestrand.no>, John C Klensin <klensin@jck.com>, Marcos Sanz <sanz@denic.de>, Jungshik SHIN (신정식) <jshin1987+w3@gmail.com>, Anne van Kesteren <annevk@annevk.nl>, "www-tag.w3.org" <www-tag@w3.org>
On Fri, Aug 23, 2013 at 12:19:02PM +0200, Mark Davis ? wrote: > registries disallow non-IDNA2008 URLs. I say URLs, because the registries > need to not only disallow them in SLDs (eg http://☃.com), they > *also*need to forbid their subregistries from having them in Nth-level > domains > (that is, disallow http://☃.blogspot.ch/ = xn--n3h.blogspot.ch). This isn't something that they do today. Indeed, there is nothing to prevent a site from putting a label there that is just the relevant raw UTF-8 bits. The thing we use to avoid this is "it doesn't work". In a different context, Dennis Jennings has been arguing for similar rules as well, and it's a mistake. We do not _want_ deep labels to have to follow the same rules as for names at the second or third levels. For instance, we want top-level domain registries to permit only LDH-labels (of some sort, including A-labels). But LDH-labels don't include underscores. Does that mean that we'd want to ban (say) SRV or DKIM TXT records? I think not. The DNS is not a global database with consistent policy. That's a deep down design feature, not a bug, and if people think that it _needs_ to have a consistent policy, then we need a different naming system. Best, A -- Andrew Sullivan ajs@anvilwalrusden.com
Received on Friday, 23 August 2013 15:23:38 UTC