On Fri, Jan 02, 2015 at 09:30:22AM -0500, John C Klensin wrote: > --On Friday, 02 January, 2015 07:57 +0100 Patrik Fältström > <paf@xxxxxxxxxx> wrote: > > > For the record, I like what is suggested the below. > > Wfm, but please note the other comments in my prior note. To > summarize, normalization is not the only issue. Even for some > normalization cases, NFC is not the cure. For example, NFKC is Indeed, which is why the proposed text also mentions confusable scripts. (I didn't include references, but they could be added, all as informative given the special circumstances here). > needed to rationalize characters of various widths but causes > problems elsewhere. So, I would suggest some additional words > that say, more or less, that, until PKCS#11 is revised to be > clear about how to handle characters outside the ASCII > repertoire, those who discover a need to use such characters > should be cautious, conservative, and expend extra effort to be > sure they know what they are doing and that failure to do so > will create both operational and security risks. I'm not opposed. I just don't think this will really add value. Thanks for your and Patrick's help BTW (my apologies if my earlier reply to you seemed caustic; I tend to get annoyed when told that the IETF "doesn't do APIs" :/). Happy new year, Nico --