symmetry between U-labels and A-labels

Erik van der Poel erikv at google.com
Sun Mar 22 18:02:59 CET 2009


Vint,

First of all, thank you for taking the time to write down the summary
of the WG status. I noticed a query that I will respond to:

> On the other hand, the symmetric relationship between the
> IDNA2008-defined A-Label and U-Label has the benefit one can use exact
> match for either U-label form or A-label forms since they are directly
> and unambiguously transformable into each other. However, this
> symmetry will not exist for cases where the IDNA2003 A-Label and
> IDNA2008 A-label for the same U-Label differ. [Query: will this be a
> material problem only for actual registrations under IDNA2003 that
> differ in A-label form from IDNA2008?]

Although it is tempting to use IDNA2008 terminology such as A-label to
describe IDNA2003 ACE labels, they are not strictly the same thing.
Currently, IDNA2008 A-labels encode Eszett, while IDNA2003 ACE labels
don't.

An important property of IDNA2008 is that there is a symmetry between
U-labels and A-labels. They can be converted back and forth many
times, without changing either of them.

An equally important property of IDNA2003 is that there is NO symmetry
between internationalized labels and ACE labels. This is because the
IDNA2003 model is that IDNA is a client-side operation only, and since
DNS server-side software does not match non-ASCII strings
case-insensitively, the client compensates for this by mapping on the
client side.

But to answer your query "will this be a material problem only for
actual registrations", no, this will be a problem wherever
internationalized labels are stored. One frequently mentioned example
of this is HTML files. One valid position to take on this is that
there are so few internationalized labels out there, that we can make
a one-time incompatible change and be done with it. Another valid
position to take is that the IETF promises interoperability in an
evolving world, and since IDNA implementations have evolved in this
way, one must take those into account.

Erik


More information about the Idna-update mailing list