Re: [sitefinder-tech-discuss] Pseudo code please

[bmanning@karoshi.com]

> Then wildcards should not be implemented in .com and .net.
> 
> 1.      The RCODE-3 / NXDOMAIN result is a useful part of the protocol
>         which many applications have come to expect in the .com and .net
>         zones.
> 
> 2.      There is no clear or compelling reason to break this functionality.
> 
> 3.      There are other ways in which a site-finder like service can be (and has
>         been) implemented that do not have this negative operational impact.
> Owen
> 
> --On Thursday, October 23, 2003 07:35:43 AM -0700 bmanning@karoshi.com wrote:
> 
> >> > The real question is:
> >> >        Given a hostname provided to my resolver library, how do I return
> >> >        a result which is different depending on whether the A record
> >> >        lookup returns an A record created by the nameservers which are
> >> >        authoritative for foo.example.com vs. an A record synthesized
> >> >        from a wildcard at a higher-level zone?
> >>
> >> Thanks, I like that description.
> >> --
> >> Kee Hinckley
> >
> > 	the answer - it is not possible for the resolver to tell.
> >
> > --bill

	perhaps - which is why the question about wildcards -AS A DELIVERY
	VEHICLE- for sightfinder is valid.

	point by point...

	1. RCODE 3 is not the only valid RCODE - per protocol.  The zone
	   admin has the right to make zone changes, including changing
	   the behaviour of the  authoritative responses.  Questions
	   about who is authorized to make such changes in .com and .net
	   are not technical.   
	2. Clear and compelling reasons for changing the zone behaviour
	   are alos not technical (at least as seen from here).
	3. If there are other technical ways to implement and deploy
	   sightfinder-like services - what would you suggest?

--bill

----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.elistx.com/unsubscribe>