Comments on draft-ietf-idnabis-defs-10

[jean-michel bernier de portzamparc]

2009/8/31 Paul Hoffman <phoffman at imc.org>

> At 4:08 AM +0200 8/31/09, jean-michel bernier de portzamparc wrote:
> >I think I would support your proposition, but I am not sure I understand
>  "A pair of A-labels MUST be compared using a case-preserving comparison.".
>
> A comparison between xn--Axdf and xn--axdf would yield "not matched".
>

Here is where I dont understand. This violates the DNS context. DNS
comparison rules MUST apply because A-labels are whole ASCII labels.


>  >Moreover, the way you phrase it seems to integrate the case-preservation
> in the protocol, i.e. in the ACE and not to keep it as a part of its use?
>
> Uppercase characters are not allowed in the U-labels anyway, so there is no
> reason to keep them in the A-labels.
>

Uppercase characters are supported but not used by ASCII Domain Name. Once
ASCII I do not see why there would be a difference based upon the origin
(how do a non-IDNA aware system would know it?).


>  >The rule proposed by Andrew Sullivan seems quite systematic and clear?
> >
> >1. The encoding of A-label1 according to [RFC3492] results in U-label1.
> >2. The decoding of U-label2 according to [RFC3492] results in A-label2.
> >3. A-label1 is equivalent to A-label2 according to DNS matching rules for
> labels.
> >4. U-label1 is bistring equivalent to U-label2.
>
> Well, no one spoke up for it. Personally, I find the introduction of
> A-label2 and U-label2 out of thin air to be confusing.
>

Well, it is the same as the present text, but clearer. It appeals to me by
its usual scientific approach.  The point is that you seem to disagree with
the rule 3. I did not realise that before using that presentation. It shows
it well.

 >We felt it addressed our point "References to the lower/uppercase image
> can be understood by DNS old-timers, but is confusing to newcomers, as it
> does not reflect the same functionality and because U-label/A-label
> lower/uppercase treatment is not the same." since everyone knows that
> punycode is case preserving.
>
> I figure it is easier to remove the comparisons to regular-style DNS
> mapping than to try to show the differences.
>

This is not where we saw the problem. We saw it in an explanation using a
non yet fully accepted image by a non-specialist. And a confusion between
upper-cases in Ascii and Unicode. We are users, not much interested in the
black-magic if we can. We just want:

(1) in real life use no more constraint IRT upper-cases than in ASCII
English.
(2) IDNA2008 is case insenstive, but we need to have it majuscule sensitive.


We have no real problem with these needs if the ASCII rule is the same in
every case, i.e. XN--ABC-DEF is the same as xn--abc-def. This is what we
understand as the protocol (on the wire) part. What the applications can do
("IDNA2008 should prohibit upper-case characters as input even though user
interfaces to applications should probably map those characters"), i.e.
"post-wire" is another different non-ACE but ACE-use story.

Or did I misunderstand the whole thing?

Portzamparc
an IETF user
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.alvestrand.no/pipermail/idna-update/attachments/20090831/648cc7d8/attachment-0001.htm