Re: [idn] iDN Transition Support Requirements

[Dan Oscarsson <Dan.Oscarsson@trab.se>]

Dave Crocker wrote:

>
>
>What portions of the Internet must support the relevant DNS mechanisms, for 
>ACE versus UTF-8?

I assume you mean the UTF-8 case where not ACE is used for backward
compatibility.


>
>Minimal use, UTF-8
>------------------
>
>a.  same as ACE
>
>b.  same as ACE
>
>c.  Every DNS component through the lookup sequence needs to support iDN, 
>before a successful lookup can be achieved.

No, you can use UTF-8 in a way that is just like the ACE handling.
Lookup by sending UTF-8 through DNS servers works fine with the current
DNS servers without change (if you use namepreped UTF-8 names just like
you do in the ACE case).

>
>d.  The target host application must support iDN, if it does any domain 
>name processing, such as to comfirm IP address and domain name matching, or 
>to display the string in native form.

The same is needed for the ACE case, the taget host must understand ACE
to display the string in native form, and must understand ACE to be able
to match a name in native form with the ACE encoded form.

>
>So...
>
>ACE permits incremental adoption.  For a first level of utility, only the 
>requesting user's software and the target host' domain server 
>administrative tools need to be enhanced.  On reflection, it will probably 
>seem obvious that one cannot get more minimal than that.

Well, if you use UTF-8 and your system uses UTF-8 as native format,
there is a good possibility that you can edit the zone files directely
using UTF-8 and start using UTF-8 without enhancing the administrative tools.
Though you probably would like to have a tool to check that the names you
enter are in the nameprepped format.

>
>By contrast, UTF-8 does not achieve minimal utility unless and until all 
>DNS-aware applications in the processing path are able to support iDN.

As indicated above, minimal usage is the same in both cases.


>My question then, are:
>
>a)  What technical errors are there in the above?

See above. You can also see in the archives for the mailing list. Not long
ago I wrote about some of the cases and the result was that the
problems were more or less equivalent.

>
>b)  What guarantees do we have about breakage/non-breakage when UTF-8 is 
>received by each of the relevant applications components, or how much 
>effort is needed obtain those guarantees?  And please note that the answer 
>to the equivalent question, for ACE, is already known:  No breakage.

ACE will give a lot of breakage. Decoded ACE names will leak into other
applications resulting in applications breaking (or just not being able
to find the host). The same will occur by using UTF-8.

No soulution is really good, both will result in a mess of host names
in ACE, UTF-8 and native forms being used in applications and protocols
which cannot handle them.

One important reason why I think people want UTF-8 is that then you can
use the same character coding on all character data and UTF-8 is the
most favoured format for interoperability today. A lot of systems work
on supporting it. ACE will always be a special DNS host name encoding.

   Dan