Re: [websec] wrt IDN processing-related security considerations for draft-ietf-websec-strict-transport-sec

[=JeffH <[e-mail]>]

Thanks for reviewing the IDNA text Pete.

 > I really, truly wish we could avoid all reference to 5895 and UTS46.
 > (And this comes from a co-author of the former.) Basically, those
 > documents are about taking user (or other sorts of "unwashed" input) and
 > doing something 'magic' before handing it to something that expects
 > proper IDNs. Really, I'd like this kind of protocol document to say,
 > "What goes is this field is something that is properly pre-processed by
 > whatever handed it to us and if it's bogus, we're not going to do
 > anything to 'help' it."

Note that that's what this spec attempts to do in the intro to section 7..

 >     This processing model assumes that the UA implements IDNA2008
 >     [RFC5890], or possibly IDNA2003 [RFC3490], as noted in Section 11
 >     "Internationalized Domain Names for Applications (IDNA): Dependency
 >     and Migration".  It also assumes that all domain names manipulated in
 >     this specification's context are already IDNA-canonicalized as
 >     outlined in Section 8 "Domain Name IDNA-Canonicalization" prior to
 >     the processing specified in this section.
 >
 >     The above assumptions mean that this processing model also
 >     specifically assumes that appropriate IDNA and Unicode validations
 >     and character list testing have occured on the domain names, in
 >     conjunction with their IDNA-canonicalization, prior to the processing
 >     specified in this section.  See the IDNA-specific security
 >     considerations in Section 13.2 "Internationalized Domain Names" for
 >     rationale and further details.

 > But I understand that this may be "happy
 > thoughts". If you really can't avoid talking about user input processing
 > in this document, I will hold my nose and say no more about it.

yeah, sigh: lacking a more generalized spec laying out all that subsumed 
detail, it seems prudent (from a completeness perspective, at least) to include 
it.  (I brought up concocting such a generalized 
how-shud-app-protocols-do-IDNA(-during-"transition-from-IDNA2003-to-2008") spec 
in the precis WG meeting in Taipei ietf-82, but there was a fair bit of 
pushback on the notion there)

But, if the ADs and IESG were to decide we could get by without such language 
(i.e. section 10 IDNA: Dependency and Migration) in this spec we could excise 
it. (I'll leave it in for now)

thanks again,

=JeffH