On Fri, Mar 06, 2009 at 02:38:37PM -0500, Edward Lewis wrote: > Remember that RFC 1123 was written in a historical epoch much different > than today. No IDNs. No URLs. No thought of domain names in > newspapers.
But depending on whether you think the text I quoted is normative, 1123 may actually forbid those IDNs at the top level. Which would be a bad thing, I think. > The BiDi issue will pretty much prevent us from seeing a TLD beginning or > ending with a digit in the global public Internet root zone. Even if the > whole issue is too new to be in an RFC (IETF's IDNABIS WG might get > around to one). Surely not. The BiDi issue may prevent anyone from seeing a U-label beginning or ending with a digit from being "added" to the root. What actually gets added to the root zone, however, is an A-label. It makes no difference whether that A-label begins or ends with a digit (it won't begin, I predict, but I'm not sure whether it might end with one), since A-labels are always ASCII. Or are you saying that display issues coming from ASCII-only labels in a BiDi display context need to govern the contents of zone files. If so, I think that really really needs to get raised in idnabis soon. That's not my understanding of the issue so far. A -- Andrew Sullivan a...@shinkuro.com Shinkuro, Inc. _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
