[DNSOP] Lame? - was Re: Asking TLD's to perform checks.

[Edward Lewis <[e-mail]>]

One of my hot buttons - meant to be one of the reasons why trying to
define "goodness" never has succeeded.  (Note, I'm using past tense.)

On 11/11/15, 1:47, "DNSOP on behalf of Viktor Dukhovni"
<[e-mail] on behalf of [e-mail]> wrote:

>  * Lame delegations are wrong.

When I did inspection of "lameness" I ran across the definition of a lame
server (in a few RFCs) being a name server, named in an NS set that
responded that it was not authoritative for the answer sought.

I cannot say that I have ever seen a definition of a lame delegation, just
a lame server.

I'll note up front that wikipedia has a page on Lame Delegation:

https://en.wikipedia.org/wiki/Lame_delegation

It references an ARIN policy from 2002:

https://www.arin.net/policy/proposals/2002_1.html

I did the research work for ARIN on that proposal.  The work did not
center on Lame Delegations but rather Lame Servers as the code was
developed.

The Wikipedia entry gives no other hint (suggested reference) as to what a
Lame Delegation is.

Other responses to a web search give other mentions of Lame Delegation,
not one reference to an RFC.

...What this message should be saying is that before re-using a known
buzzword, please offer a precise definition of it.  I recognize a
hand-waving feel for what is a lame delegation, but with that as a
starting point, you'll never come to consensus on a testing tool.