I would have agreed with you, but the text in the latest (-06) draft
looks good too.
FWIW, I think the addition of an internationalization considerations section
to the base spec also makes sense. The one I wrote only deals with issues
specific to vacation (and maybe refuse and notify).
The only typo I noted:
> The Sieve language itself allows a vacation response to selected
> based on the content of the original message.
needs a "be" -- "to be selected" (not "to selected")
Thanks for the catch - Kristin Hubner also pointed this out in private
email. Unfortunately I made the change to the XML but then forgot to rerun
xml2rfc one last time before submission. I'll fix it during auth48.