Suggested resolution for the @domains attribute kerfluffle

[Kristen James Eberlein <kris@eberleinconsulting.com>]

After a lot of reading and re-reading of the spec topics, the only thing 
that is clear to me is that --like many DITA 1.2 items -- there is a 
lack of clarity as to how @domains entries should be defined for 
structural specializations.

To move forward with the least impact and disruption, I suggest that the 
specification be updated to explicitly allow the (topic) and (map) 
tokens as a way to indicate the base DITA vocabulary for a specific 
document type shell.

I do not think that we should use normative language in the update. We 
cannot use a MUST statement, and I do not think that even a SHOULD 
statement is warranted, because it is of little enough practical value 
that I don't think there is any benefit in updating 1.1 or 1.2 compliant 
shells to add this new token.

However, this update will offer a solution to the issue that Robert 
originally raised of a document-type shell that does not have any 
domains integrated.

--
Best,
Kris

Kristen James Eberlein
Chair, OASIS DITA Technical Committee
Principal consultant, Eberlein Consulting
www.eberleinconsulting.com
+1 919 682-2290; kriseberlein (skype)