On 14/10/2020 06:33, Tom Honermann wrote:
> Thank you, Nial. I'll ask for a volunteer at the SG16 telecon scheduled
> for this week (in under 24 hours) to perform a review. Unless that
> review raises new SG16 specific concerns, I'm not inclined to spend
> further SG16 telecon time on this paper as I think LEWG is well
> positioned to take it from here. LEWG and/or LWG can always request
> specific input from SG16 if their respective chairs feel doing so is
> warranted.

Billy gave extensive, highly transforming, "only an implementer would
see this stuff" feedback which widely transforms the paper. Basically it
has to be rewritten, and I'm about half done in that rewrite. I've also
merged feedback from Corentin, and I think a better proposed wording has
resulted.

The reference implementation in LLFIO has been upgraded to match the
future wording, and source compatibility breakage has not been terrible.

Please do seek the volunteer however, and I'll get them draft 2 by next
week. The biggest change in R4 relevant to SG16 is configurable
separator interpretation, where we shall need to take a decision about
how it needs to interact with Unicode normalisation and separator
collapsing. Earlier editions did not have it at all, LEWG (via a comment
from Billy actually) added it in Prague.

Thanks for your help.

Niall