Yeah, there's no reason to use U-labels here. Let's just require that the names be in ASCII. I don't think there's really a need to require anything further than that (servers can do case folding).
Note that this only affects Authorization.identifier.value for Authorization.identifier.type == "dns". The names in the CSR can still have all the delightful variation that X.509 allows. On Mon, Aug 29, 2016 at 7:21 PM, Philipp Junghannß <[email protected] > wrote: > shouldnt IDNs just get Punycoded? at least that's the common standard for > IDN stuff iirc. > > > 2016-08-30 0:51 GMT+02:00 Roland Bracewell Shoemaker < > [email protected]>: > >> I'd be interested in hearing peoples thoughts about this PR which adds >> language about how IDNs should be encoded by clients who wish to use >> them as identifier values. >> >> https://github.com/ietf-wg-acme/acme/pull/184 >> >> _______________________________________________ >> Acme mailing list >> [email protected] >> https://www.ietf.org/mailman/listinfo/acme >> > > > _______________________________________________ > Acme mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/acme > >
_______________________________________________ Acme mailing list [email protected] https://www.ietf.org/mailman/listinfo/acme
