Germany made an interesting proposal:
DE - 148
Given the scope, breadth and depth of the specification, including the numerous passages marked as "informative" only, the conformance criteria here seem few and vague.
Specify exact conformance criteria, possibly for different levels (minimal, full etc.) and/or for specific subsets of the specification and/or from the perspective of different categories of implementing software (visual editors, renderers, batch filters, pass-through systems etc.)
f.e.g. Profiling could be a proper solutionPart 1, section 2.5
ed
http://www.dis29500.org/de-148/
While it is marked as an editorial proposal it has technical implications which can be interpreted as similar to the French proposal to split Open XML into a core part and an extension part. The idea to specify conformity levels of the specification should be more discussed. In certain use cases it might be sufficient to implement only portions of the DIS 29500 standard that is extraordinary large (as you know).
Also regarding the coverage of incomplete implementations with potential patent licensing and a possible additional inclusion of the Microsoft BIFF specification into the standard a model with conformance levels could help to structure and prioritize the standard.
I am curious about your opinions on that German proposal.