Shepherd writeup

Shepherd writeup for draft-ietf-precis-mappings-11

1. Summary

The document shepherd is Peter Saint-Andre.

The responsible Area Director is Barry Leiba.

This document provides guidelines for supplemental and alternative
handling of some Unicode characters in the context of PRECIS (RFC 7564).
The supplemental handling covers mapping of protocol-specific delimiters 
(e.g., "@") and special characters (e.g., mapping of non-ASCII space to 
ASCII space). The alternative handling covers language-specific mapping
of uppercase characters to lowercase characters (e.g., Turkish dotless
i) as an alternative to Unicode Default Case Folding, along with one 
instance of context-specific mapping (Greek final sigma).

The requested publication type is Informational because the document
provides non-normative guidance for protocol designers and application

2. Review and Consensus

Internationalization is full of messy complexity. The topic of case
mapping, although it might seem straightforward to English speakers, 
can be far from clear-cut. Mapping of uppercase characters to lowercase
characters sometimes varies by language (e.g., certain characters are 
handled differently in Turkish and Lithuanian than they are in other
languages that use the same characters) and by geographical location 
(e.g., the French language as written in Canada as opposed to France).
PRECIS-based software implementations do not necessarily have access 
to information about a user's language or locale. Furthermore, as a 
generalized technology for internationalization, PRECIS does not include 
a way to negotiate such parameters. 

In the main it is safest to use Unicode Default Case Folding for case 
mapping, as defined in the Unicode Standard and as recommended by the 
PRECIS framework (RFC 7564). However, where information about a user's 
language or locale (or the context in which a particular character is 
used) is available, it can make sense for an application protocol or 
implementation to adjust its handling of case mapping based on such

Because of the complexities involved, the sense of the PRECIS WG was
that it could not provide normative recommendations for case mapping 
beyond those defined in the Unicode Default Case Folding algorithm. 
Nevertheless, the working group saw value in providing informational 
guidance regarding language-, locale-, and context-dependent mappings.
This is similar to the relationship between the core Internationalized 
Domain Names in Applications (IDNA) protocol and the informational 
guidance provided in RFC 5895. 

In contrast to the topic of case mapping, the guidelines regarding
delimiter mapping and special mapping were uncontroversial and elicited
no significant concern among working group participants.

3. Intellectual Property

Both authors have confirmed that they are not personally aware of any
IPR related to this specification.

4. Other Points

The document makes normative references to the Unicode Standard, but
that is not problematic because such references are common in IETF
documents (and this document is Informational anyway).

This document requests no actions by the IANA, does not require expert
review, and needs no other special handling.