consensus item - "IDNA2008"

Erik van der Poel erikv at google.com
Mon Dec 22 00:15:53 CET 2008


Hi Paul,

First you agreed that we should focus on more substantive issues than
the name "IDNA2008", but then, while explaining your preference, you
discussed the rather more substantive issue of compatibility with
IDNA2003. I just thought that was an interesting flip. :-)

Anyway, we often talk about changing or switching the prefix (xn--),
but we shouldn't kid ourselves: if we introduce another prefix, we
would be *adding* it, and probably supporting both prefixes for quite
a long time, if not forever. I agree that another prefix is the only
way to guarantee compatibility with IDNA2003 pre-processors.

But I believe we have a wide range of opinions about how compatible
IDNA200X must be. Some believe in strict compatibility with IDNA2003,
others believe in the viability of the current IDNA200X drafts. Yet
others might fall somewhere between these extremes. The coming WG Last
Call, IETF Last Call, IESG debate, discussions among implementers and
decisions by implementers will certainly be interesting.

Erik

On Sun, Dec 21, 2008 at 10:00 AM, Paul Hoffman <phoffman at imc.org> wrote:
> At 12:41 PM -0500 12/21/08, John C Klensin wrote:
>>Paul, with the understanding that I can live with it either way
>>--and that the suggestion was an attempt to see if we could
>>focus on getting the more substantive issues resolved before
>>spending too much more time on editorial details-- I'd like to
>>understand your reasoning.   I think Vint's suggestion (perhaps
>>modulo the comment in my note) is reasonable.  Do you see a need
>>to have a last-minute debate on this if we can reach agreement?
>>If so, why?
>
> Because a number of people expressed a preference for something other than IDNA2008, and because I fully agree with you that we should focus on the more substantiative issues first.
>
>>If we cannot quickly reach agreement, I still prefer "wait until
>>we are nearly finished".  But I don't see a problem with asking
>>the question.
>
> Fine. So, now we will have the discussion that you didn't want to have.
>
> I disagree with calling it IDNA2008 because it was not completed in 2008. We don't even know if it will be complete at all: we have not had a WG last call, much less an IETF last call or an IESG debate. I am far from convinced that the IESG is going to approve of what we are doing, particularly the "allow seriously incompatible changes to the protocol without changing the version string so that old clients will not interoperate". We in this WG have agreed with that direction, but the IETF has not, and the IESG has not.
>
> The reason I don't want to propose "IDNA2009" is that we might end up with "IDNA2010" or, unfortunately, "IDNA2003". I think your suggestion of having this discussion when we know what the protocol will be was a good one.
> _______________________________________________
> Idna-update mailing list
> Idna-update at alvestrand.no
> http://www.alvestrand.no/mailman/listinfo/idna-update
>


More information about the Idna-update mailing list