IAB Statement on Identifiers and Unicode 7.0.0

Paul Hoffman paul.hoffman at vpnc.org
Wed Jan 28 02:49:22 CET 2015

On Jan 27, 2015, at 5:13 PM, Shawn Steele <shawn.steele at microsoft.com> wrote:
>> I also wonder why you think the statement is "overkill".  We have some cases that we definitely believe are going to be troublesome, and we recommend that people avoid using them in new identifiers until that issue is sorted out.  We have a recommendation that we not make the structural problem in IDNA worse.  And we exhort the IETF to do something.  Where's the overkill?
> The WG was already discussing it :)  

Which WG? If you mean this mailing list: this list is *not* an IETF WG at all. It is just a mailing list, with no chair and no consensus process.

If you mean some UTC WG, well, most of us wouldn't know about that.

> It focuses on edge cases of confusable characters.  These are a very small part of the potential for confusion in IDNA.  

We are not trying to deal with all confusion in IDNA or TUS. IDNA2008, for better or worse, made protocol design decisions based on verbal assurances from TUC about how composed and decomposed characters would be treated going forwards. It appears that the rules have now changed, and the IETF needs to determine what protocol changes it needs to make.

> The key part is perhaps: "The IAB does not dispute the reasoning that the Unicode Technical Committee uses for such determinations (neither is the IAB qualified to dispute that reasoning). The IAB’s sole concern in this case is whether the correct choice of character will be discernible in the context of identifiers."
> Certainly having identifiers that are consistent is good, however we already have PVALID characters that are confusing, and they don't even have to look exactly the same.  (naïve/naive for example, or fußball/fussball).  Drilling down on a few specific characters seems to me to ignore the bigger problem of confusable identifiers.  IMO if the wider problem was addressed, then these cases would be fairly uninteresting.  It's like we can't see the forest for all the trees.

There is no sane way to address the wider problem: we have proven that twice. Given that, we need concrete examples of problems and solutions.

--Paul Hoffman

More information about the Idna-update mailing list