IETF Last Call Review of draft-davids-forsalereg-18
review-davids-forsalereg-18-artart-lc-nemoto-2025-12-28-00

I am the assigned ART-ART reviewer for this draft.

Summary:

  This draft is clear and easy to read, and the overall approach of signaling
  domain availability via a reserved underscored DNS leaf node fits well with
  existing DNS usage patterns. From an ARTART perspective, the document is
  generally in good shape. A small number of clarifications would further
  improve implementability at the application layer and reduce the risk of
  divergent interpretations.

Major issues:

  None.

Minor issues:

  Section 3.2 discusses the handling of TXT RDATA containing UTF-8 text,
  including the representation of Internationalized Domain Names (IDNs) as
  either A-labels or U-labels. While the intent is understandable, it is not
  entirely clear whether U-labels are intended only for textual or documentary
  representation, or whether applications are expected to process them directly.

  A brief clarification of the expected roles of A-labels versus U-labels,
  particularly with respect to operational DNS resolution versus presentation
  or informational use, would be helpful. For example, a short statement
  indicating that DNS resolution relies on A-labels while U-labels are intended
  for presentation would remove potential ambiguity for application developers.

  Additionally, the document states that UTF-8 data may be encoded using "octal
  escapes". In the DNS context defined by RFC 1035, the \DDD escape form
  represents decimal-encoded octet values rather than octal values. Using the
  term "octal escapes" here may be misleading and could result in incorrect
  implementations. Consider rephrasing this to explicitly refer to decimal
  numeric escape sequences or to RFC 1035 character-string encoding.

Nits:

  None.

Additional Note:

  As a purely optional and non-blocking observation, and speaking from a
  Japanese language perspective: in the example in Section 3.2,

      _for-sale IN TXT "v=FORSALE1;ftxt=\229\163\178\229\135\186"

  the UTF-8 encoded text decodes to the string "売出". While this form is
  understandable, when expressing a state in Japanese it is more common to use
  the form "売出中" (meaning "now for sale") rather than "売出" alone, although it
  would increase the encoded length to nine octets
  (\229\163\178\229\135\186\228\184\173). This comment is offered only as a
  linguistic note and does not require any change to the document.