Comments on

[Arthur Ryman <arthur.ryman@gmail.com>]

I reviewed the current version of [1] and have some comments:

- overall, the document is coming together well

- there are minor grammar errors and typos. these should be fixed in a
final editorial pass prior to publication of the working draft

- Fig. 2 and the preceding paragraph are duplicates of Fig. 1 and Section 1.2.

- The table in Fig. 6 needs a more detailed mapping to the PROMCODE
domain model. Here I assume that the Planned values for the metrics
map to Target resources.

- in section 4.4, Fig. 7 I suggest that Target be a subresource of
Artifact and that Measure be a subresource of Target and Measurement.
This eliminates TargetContainer and MeasureContainer

- there are some inconsistencies between the resource definitions in
5.1 and the vocabulary in 5.2 (e.g. in the use of ems: terms). Both of
these sections should be automatically generated from Turtle files to
improve consistency.

- I suggest that we use Turtle for all RDF examples since it is the
most readable format. The examples in section 7 should be converted to
Turtle.

- The examples should be more complete and the text should refer to
them and describe them.

- The Excel examples should have corresponding Turtle examples.

- The spec does not define how to map Excel to RDF, so these mapping
examples should be removed (e.g. Fig. 13). The mapping is an internal
implementation detail of the Excel client.