ietf-smime
[Top] [All Lists]

Re: UTF8 vs. Punycode

2007-08-14 15:58:50

At 12:05 AM +0200 8/15/07, Simon Josefsson wrote:
Paul Hoffman <paul(_dot_)hoffman(_at_)vpnc(_dot_)org> writes:

 At 11:30 AM +0200 8/14/07, Simon Josefsson wrote:
One risk is that the specification cannot use Unicode code points from a
newer Unicode version than IDNA ToASCII supports, right now that means
Unicode 3.2.

 That is not necessarily true. The current version of IDNA supports
 Unicode version 3.2. A future version of IDNA may support later
 versions of Unicode.

I believe the assumption here is that the document would reference the
current IDNA version.  Or are you suggesting that this document should
wait until the IDNAbis effort is done?

The former. But that does not change what I said: a later version of IDNA that has the additional characters you were talking about could be replaced in the spec with near-zero effort.

 >>Some IETF protocols can easily negotiate support for UTF-8 on
both sides, and using UTF-8 rather than Punycode seems more robust and
like better engineering to me.

 Un-normalized fails miserably when exact matching is needed, such as
 it is in IBE.

Sending UTF-8 does not preclude normalizing from happening when it is
needed.

Ah, I misunderstood your proposal. I thought you meant "send UTF-8", not "do some normalization that is not specified here, then send UTF-8". It would be good if you could give your complete proposal so people can compare them fully.

--Paul Hoffman, Director
--VPN Consortium

<Prev in Thread] Current Thread [Next in Thread>