Proposed document structure and content for IDNA

[Paul Hoffman]

Greetings again. This is in reply to many comments that were made at 
the face-to-face meeting in Philly meeting, and prompted by Vint's 
message yesterday about preparing for the Dublin meeting.

There are three large problems with the structure and content of the 
current set of documents; fortunately, all are easy to fix at the 
same time.

- An implementer would have to read some parts of all four documents 
in order to do any implementation of IDNA2008

- An implementer would need to determine which parts of the four 
documents were commentary / history / rationale versus which parts 
are needed for an interoperable implementation

- There is a possibly-permanent lack of consensus about the rationale 
for updating IDNA2003 to IDNA2008

Given these three, and easy solution would be to create a single 
document that has just the needed protocol pieces without any 
personal commentary, history, or rationale.

Having a single document such as this would make it much easier for 
the WG to review. The long, pendulous threads from the past few 
months have often been more about rationale than about protocol, and 
even when they were about protocol, they devolved into rationale 
arguments.

Having a single document such as this would make it more likely that 
IDNA2008 implementations would fully interoperate. Implementers would 
not need to guess at the WG's intentions: the protocol would be fully 
specified in one place.

Having a single document such as this would make IDNA2008 clearer 
than IDNA2003. Our decision to split IDNA2003 into multiple 
interlinked documents seemed right at the time, but we have heard 
repeatedly since then that it wasn't a good idea. Including some of 
or rationale also seemed like a good idea, but that turned out wrong 
as well.

In summary, we should:

- Merge the four documents into one (keeping all the authors, of course)
- Completely strip out the commentary / history / rationale
- Move forwards with one concise protocol document

Thoughts?