At 10:31 -0800 1/28/12, todd glassey wrote:
In the interim we think a COPYRIGHT USE STATEMENT published as a TEXT
type record may work for systems which use legacy packages to operate.
Do you mean like the response to these queries?
$ dig ac. txt +short
or
$ dig at. txt +short
and a handful of others?
Beyond that, there is a process defined to add a new DNS RR type defined in BCP
42.
What we came to is that the lookup process may need to be amended to
support copyright disclosure for the records responded and that this is
key to global copyright protection.
As far as altering the query-response mechanism to always include a
copyright, the workload amounts to doing DNSSEC all over again.
Adding a copyright to every response is akin to adding an digital
signature to every response (set). Even though adding the text would
be less of a strain than computing the signatures, alterations are
needed to the basic algorithms (like my fav RFC 1034/sec 4.3.2 +
updates) which in turn means new code all around.
(Starting with the 2004 release of the DNSSEC specs, it took about 6
years to get the root zone signed. Imagine trying to decide who the
copyright holder is for the root zone!)
This wouldn't be easy because the DNS just wasn't designed with legal
restrictions in mind. Adding a set of requirements means some amount
of re-engineering. And I'd add that this set of requirements might
not be seen as universal.
What I'm suggesting is that beyond adding TXT records in the apex of
the zone defining anything is going to be a long hard road.
--
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Edward Lewis
NeuStarYou can leave a voice message at +1-571-434-5468
2012...time to reuse those 1984 calendars!
___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop
