[DNSOP] Re: [EXTERNAL] Re: Call for Adoption: draft-davies-internal-tld

[David Conrad <[e-mail]>]

Joe,

On Apr 17, 2025, at 12:39 AM, Joe Abley <[e-mail]> wrote:
> We should not need TLD-specific handling. TLDs in general are and should not be special.

That far off bump on the horizon you see is the ass end of a ship that sailed long ago.

As you’re no doubt aware, there are 7 TLDs currently listed in https://www.iana.org/assignments/special-use-domain-names/special-use-domain-names.xhtml. Each of those can imply (but do not necessarily require) special handling, either in authoritative code, resolver code, or the surrounding client or server infrastructure. Given those seven, it seems silly to me to not add another TLD that may imply special handling by DNS developers or operators.

> The root server system is not the part of the DNS infrastructure we need to worry about.

Heh. If this were only true, the decade-long effort to develop a governance structure for the RSS could be wound down and Jim, Geoff, and others might be slightly less grumpy. However, I’d agree that the impact of .internal to the RSS will not be significant, regardless of whether .internal is documented with the other seven special use TLDs. 

> All the tiny exceptions and niggly extra considerations add up on top of a protocol that is already unwieldy and undocumented.

And I see this draft as part of an effort to reduce the last bit.

> Every little extra nibble from the duck moves us closer to mortal injury. This is Bert's camel; the additional burden is tiny and insignificant until the camel dies.

This is overly simplistic. Given the evolution of the Internet and the use of the DNS, the DNS will need to be modified. Every modification to the DNS implies (or at least should) a cost/benefit analysis. In this particular case, what I see is being proposed is the addition of an entry to the SUDN, presumably with a pointer to a document that explains why .internal is special. From my perspective, the benefit is consistency and helping DNS implementers understand the protocol and the cost is a document somewhere authoritative that says “for internal use only.” Do you see other costs?

Regards,
-drc