Re: [sitefinder-tech-discuss] followup from 15oct secsac

[Owen DeLong <owen@delong.com>]

--On Thursday, October 16, 2003 9:01 AM -0700 bmanning@karoshi.com wrote:

> > > 	on delivery:
> > > 		wildcard entries come in a couple flavors.
> > > 		using wildards for TXT and MX are fairly benign.
> >
> > Wildcard MX are very MALIGNANT unless they point to a name that returns
> > NXDOMAIN and there is no A record.
>
> 	say what?  why oh why do you think that people use wildcard
> 	MX records in this fashion?  It appears you are getting tangled
> 	back up in the delivery method vs service.  granted MX rrs
> 	(wildcard or not) imply something at the other end beinging
> 	able to deal -correctly- with what is presented on port 25.
Right.  An MX record implies a MAIL EXCHANGER at the other end that agrees
to correctly handle Email for all domains for which it is listed as the MX.
An MX record that points to a server which does not is by definition 
MALIGNANT.

> 	your still hunting for that old RCODE-3 stuff...
> 	what -IF- one of the design goals for this goofy  idea
> 	was to -ELIMINATE- all RCODE-3 replies?
Then, it is a bad goal, which, when presented to IETF properly would most
likely be voted down in a dizzying display of speed never before seen in
the IETF.

> 	in any case... how would folks feel about something like this:
>
> 	com.  in soa (bits'o'rr)
> 	      in ns  lumpy.
> 	      in ns  gravy.
> 	*     in naptr (naptr-bits)
>
> 	whango z tango --- no pesky A rr... :)
If the wildcard goes up against a new RR type, and, there is a fair process
by which any entity who wishes to do so can be added to the NAPTR records
for these TLDs, then, I have no problem with it.  The registry should be
prohibited from any ownership in any target of NAPTR records.  Existing
RR types should not be wildcarded in .net or .com.

> > > 	on service:
> > > 		what will be offered and why?
> > > 		what will not be offered and why?
> > >
> > Very good questions, but, the issue remains that there is no
> > implementation of these wildcards that provides a benefit in excess of
> > the damage.  There is NO reason (other than Verisign's greed) to turn
> > these wildcards back on.
>
> 	that whole downside thing that you elided.
> 	
Yep.

> > Owen



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