[Idna-arabicscript] mapping of Full Stops

[Sarmad Hussain]

John, thank you for referring to the visual similarity to hyphen.  This creates an issue with U+06D4 (though evident only when one argues not collapsing it with hyphen):

	If U+06D4 is allowed, the domain name mail.mydomain.com.pk 
	can also be represented as mail۔mydomain۔com۔pk (all Arabic full stops).
	However, this may be phished by mail-mydomain۔com۔pk (first hyphen, 
	then Arabic full stops), which would take the user to mail-mydomain 
	(with hyphen) instead of mydomain within .com.pk.  

In this context, I agree with the group that this needs to be investigated further and would retract the request from the Last Call.

However, do note, in context of other comments, U+06D4 is a DISALLOWED code in IDNA200x protocol.  This does imply that any previous registered domain names with U+06D4 as part of the label would be rejected by any application conforming with the upcoming IDNA protocol

Thank you all for the fruitful discussion.  

Regards,
Sarmad

 

> -----Original Message-----
> From: John C Klensin [mailto:klensin at jck.com]
> Sent: Tuesday, October 13, 2009 3:45 AM
> To: Wil Tan; Erik van der Poel
> Cc: Andrew Sullivan; sarmad.hussain at nu.edu.pk; idna-
> update at alvestrand.no; Martin J. Dürst; Vint Cerf; Arabic Scripts IDNA
> Subject: Re: [Idna-arabicscript] mapping of Full Stops
> 
> 
> 
> --On Tuesday, October 13, 2009 08:42 +1100 Wil Tan
> <wil at cloudregistry.net> wrote:
> 
> > xn--klb.com (U+06D4 + ".com") is registered and resolves.
> 
> In other words, substituting a look-alike glyph, we have a
> domain name that would appear to be
>       -.com
> we could also have U+06D4 U+002D U+06D4 and see
>       ---.com
> (in some fonts, the hyphen would be rendered a little higher
> than the Arabic full stop)
> 
> I think think this illustrates several things, one of which
> involves some rude words about both a certain registry and
> ICANN's ability and willingness to enforce its guidelines.
> 
> In a different world (something I wouldn't suggest and the fact
> that we are in Last Call has nothing to do with it), it would be
> reasonable for us to be having a discussion, not about using
> U+06D4 as a label separator (i.e., treating it is "dot"), but
> about mapping it to hyphen-minus as a separator character inside
> labels.
> 
>    john
> 
> 
> 
> 
> __________ Information from ESET NOD32 Antivirus, version of virus
> signature database 3811 (20090129) __________
> 
> The message was checked by ESET NOD32 Antivirus.
> 
> http://www.eset.com
> 
 

__________ Information from ESET NOD32 Antivirus, version of virus signature database 3811 (20090129) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com