[regext] AD Evaluation: draft-ietf-regext-epp-delete-bcp-08

[Orie Steele <[e-mail]>]

# Orie Steele, ART AD, comments for draft-ietf-regext-epp-delete-bcp-08
CC @OR13

* line numbers:
  -
https://author-tools.ietf.org/api/idnits?url=https://www.ietf.org/archive/id/draft-ietf-regext-epp-delete-bcp-08.txt&submitcheck=True

* comment syntax:
  - https://github.com/mnot/ietf-comments/blob/main/format.md

I have only minor comments, and I believe this document is ready for IETF
Last Call.

## Comments

Thanks to Andy Newton for the shepherd writeup.

Highlighting a comment from the writeup:

```
Additionally, there were discussion regarding the use of BCPs to suggest
practices that are not know to be practiced. In other words, how can a BCP
normatively
require a practice that has not been observed because it is not a "current"
practice.
It was noted that RFC 2026 permits BCPs to describe what is believed to be
the best
way to perform some operations or IETF process functions.
```

Indeed the document highlights several patterns which would "not be best",
and provides recommendations for what is believed to be best at this time.

### Why not MUST?

```
414   The service SHOULD provide responses that indicate problems with a
415   domain's delegation, such as non-existence or include controlled
416   interruption IP addresses [RFC8023].  This practice has been observed
417   in use.
```

Under which circumstances can this should be safely ignored?

### Use of .invalid recommended?

.alt is clearly forbidden:

```
440   Clients may rename host objects to use ".alt" or another non-DNS
441   pseudo-TLD as suggested in [risky-bizness-irtf].  This practice has
442   not been observed in use.  This practice MUST NOT be used.
```

So I wonder if there should be clear guidance on treatment of .invalid.

```
471   Clients may rename host objects to use ".invalid" or another reserved
472   special-use ([RFC6761]) TLD as suggested in [risky-bizness].
```

Because the comments on .alt come before the comments on .invalid, I get
the feeling that use of .invalid is encouraged... even though there is no
explicit guidance.

Later:

```
748      use domain issues described in [RFC8244].  Use of
749      "sacrificial.invalid" (see 5.1.4.3) as the parent domain for the
750      host objects is RECOMMENDED to avoid the overhead of creating a
751      new TLD using either IETF or ICANN processes that offers no
752      additional operational benefit.
```

Perhaps add a brief comment and a pointer to this recommendation when the
concept is introduced.

-- 


OS, ART AD