In article <fdrn6r$5d2$1@atlantis.news.tpi.pl>,
flawius <flawius@gmx.net> wrote:
>Rod Dorman wrote:
>> In article <fdr86k$br7$1@atlantis.news.tpi.pl>,
>> flawius <flawius@gmx.net> wrote:
>>> ...
>>> but this information is not authoritative. Right?
>>
>> Assuming they set it up correctly, it *is* authoritative. There might
>> not be any RRs for hosts yet but it will answer authoritatively that
>> foo.example.com doesn't exist.
>
>I thought that NS RRs residing on the com domain nameservers (saying
>that the namesrevers for example.com are ns1.aregistrar.net and
>ns2.aregistrar.net) are not authoritative and that authoritative NS RRs
>are those residing on the example.com domain namesvers (that is
>ns1.aregistrar.net and ns2.aregistrar.net) because of:

I think your confusion might be my use of just the word
'authoritative' instead of saying authoritative for example.com.

Just saying a nameserver is authoritative doesn't give a complete
picture. One should really include *what* its authoritative for.

For example, the . (root) nameservers are authoritative for all the
TLDs. The .com nameservers are authoritative for all the <domain>.com
domains. The example.com nameservers (ns1 and ns2.aregistrar.net in
your example) are authoritative for example.com.

--
-- Rod --
rodd(at)polylogics(dot)com