On Feb 22, 2015, at 9:12 AM, Måns Nilsson <[email protected]> wrote:
>>> Anything in STD80?  That's kinda strange.  The STD13 "rule" is
>>> effectively the Preferred Syntax, which is LDH.

I thought this was applicable to "hostnames", not "domain names".

>>> Alternatively, labels are octets.

Well, 0 to 63 octets.

>>> Note that there are most definitely systems that put
>>> UTF-8 in zones, and this is entirely fine and legitimate;

Yes, to wit:

http://www.thedomains.com/2015/02/20/coca-cola-puts-smiley-face-emojis-into-web-address-using-ws-domains/

>>> it just doesn't work that well on the Internet maximally construed.

Somewhat as an aside, I suspect the desire to ensure this is what is leading to 
ossification.  Not a criticism, just an observation.

> And since that is the case, and since there most likely are zone file
> parsers that barf on anything !ASCII I still believe it is wise to
> limit the standard file format.

As far as I can tell, there are zone file parsers that barf on pretty much 
anything (including 20+ year old RR type names).

> As I understand it, one of the goals
> with defining the file format is that a file conforming to the format
> should pass most parsers today.

While I appreciate the pragmatism, I'm not sure I see the value of this 
approach given (from past experience, but have little current data) few zone 
file parsers actually do the right thing in all cases. I'd imagine the goal 
here is aspirational: what does the DNS technical community think the concise 
and comprehensive zone file format _should_ be, regardless of the abominations 
found in the myriad implementations.

Regards,
-drc

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to