Talk:European Data Point Methodology V2.0
From XBRLWiki
| Revision as of 07:09, 10 October 2013 (edit) Anna-Maria.Weber (Talk | contribs) (→Comment-15) ← Previous diff |
Revision as of 07:10, 10 October 2013 (edit) Anna-Maria.Weber (Talk | contribs) (→Comments) Next diff → |
Revision as of 07:10, 10 October 2013
Comments
Comment-01
[TD] Or better "Introduction" for section title? (comment "b" in the file Comments about CEN/TC XBRL)
Comment accepted. TD to adapt
Comment-02
[TD] Ignacio suggests "In the page 7, figure: Figure 1 —Structural Perspective, the cardinality of 1 is not necessary. [TD: I can not change the graphics]
Comment rejected.
Comment-03
[TD] Ignacio suggests here to repleace the sentence above the comment with: "In the DataPointModel a Hierarchy forms are sets of concepts of a domain (DefinedMembers) of a dimension (EnumerableDimension) arranged in a hierarchical disposition." Myself I note a problem with the expression: "a Hierarchy forms are sets" (should it bee "a Hierarchy forms sets ..."?
Comment accepted: "are sets" -> "a set" TD to adapt.
Comment-04
[TD] Suggestion by I. replace "A Module is a group of DataCubes that carry relevant identical semantics and may serve the reporting process" with "A Module is a group of DataPoints with its appropriate Dimensions and concepts of the dimension (DataCubes) ...."
Comment to be further processed: make sure that ref links are present in the Word Document.
Comment-05
[TD] Suggestion by I.: “4.3.10 Dimension, …. A Dimension can refer to a Domain. ….”. I would exchange by in “4.3.10 Dimension, …. A Dimension must refer to a Domain. ….”.
Comment rejected.
Comment-06
[TD] Comment by I. :“4.6 Hierarchical Perspective …. 1) When using multiple DimensionedElements on a single Dimension that has a Hierarchy in its DefinedMembers, the required math may not be possible to perform.”. I don’t understand. Only, it is possible to define a data point with only a member domain by dimension. Is this that you want to say?
Comment accepted. Roland to edit the segment.
Comment-07
[TD] Comment by I. : h. In page 18, “4.6.2 RuleRelationship … The list of possible signs in a DataPointModel is not determined. Examples are: + (plus sign) or - (minus sign). ….”. Can have in a hierarchy “*” or “/”? I think: “4.6.2 RuleRelationship … The list of possible signs in a DataPointModel must be “+” or “-“. Examples are: + (plus sign) or - (minus sign). ….”.
[TD] But I think this comment refers to a previous version of the document.
Comment reject.
Comment-08
[TD] comment by I.: figure 5, the arrow to tablesheet is longer. [TD: I can not modify the graphics
Comment accepted. Katrin to modifiy.
Comment-09
it is a doubt. “Rule 1.9 — There MUST NOT be a doubling of DefinedMembers in the same Dimension. A DefinedMember MUST only be references once in a Dimension.”. I understand that a member domain can belong to several dimensions, cannot it?
Comment rejected.
Comment-10
“In general” should be added before parameters. The rest like Booleans or strings are exceptions. Multidimensional models are in general oriented to analysis.
Comment-11
Each DataPoint MUST be represented in one DataCube
Suggestion: Replace /one/ by /exactly one/ (not one or more)
KH: Same DataPoint could be used in COREP and FINREP.
Comment-12
BdF:"“A DataCube must be part of at least one Module” This rule forbids a modelled information that is not yet in production." RH:"Indeed, is that a requirement? Models should be able to contain information that is not (yet/anymore) used?" KH: "The currency period rules the begin of production."
Comment-13
Suggestion by BdF: "The business template is to be split in two or more tables to prevent that the same Dimension is associated to more than one axis"
RH: Hmmm. It's not about separating the dim from X and Y, but from X1 and X2.
Comment-14
BdF: "I think in DPM, a domain can reference several dimensions"
RH: In XBRL, yes. In (Victors) DPM, no. KH: The link in the graph is unidirectional. The domain does not know which dimension refer to it.
Comment-15
BdF:" A dictionaryElement can be owned by several taxonomy" RH: Not in the setup where all elements are created only once. Either re-use the schema or use versioning. KH: What about metrics that are defined for COREP and FINREP
