> > Hi, > > Generally, the document reads well and is quite concise which I think is good. > Below are few comments. > > - The notion of Topbundle is put forward in few places as the top > level non terminal expression. I am not sure if that serves a > purpose, specially that a top level bundle may, I imagine, be > references by other non terminal expressions, e.g., wasAttributedTo. > > - For the reader to understand Example 2, we may need to add a few > words to say what a, g2 and u1 represent, which can be done in one > sentence I think. For example, "where a is the activity the used e1 > to generate e2. u1 and g2 represents the usage and generation, > respectively." > > - In Section 2.5: "cooments" -> "comments" fixed > > - In Section, 3.1.2, I suspect that "time" is non terminal, and > therefore, the reader may expect it to be broken down into terminals > terms. I know this is beyond the scope of PROV-DM, but it would be > good to add a sentence saying that. Time is defined in section 3.7.3.2 as the DATETIME terminal, and the definition can be found in xml schema. [49] time ::= DATETIME > > - There are two terms that are used to define identfiers: "Identifier" > and "OptionalIdentifier". Why not use "Identifier" and when it is > optional use "(Identifier)?" as specified in Section 2.2? It's production [10], which also include ";" [10] optionalIdentifier ::= ( identifierOrMarker ";" )? > > - In section 3.1.4, it is said that "Even though the production > usageExpression allows for expressions used(a2, -, -) and used(-; > e2, -, -), these expressions are not valid in PROV-N, since at least > one of id, entity, time, and attributes must be present." However, > according to the definition of Usage in [12], the activity > identifier is not optional. correct, the activity identifier is not optional > > - Section 3.2.1, defines derivation as: [17] derivationExpression ::= > "wasDerivedFrom" "(" optionalIdentifier eIdentifier "," eIdentifier ( > "," aIdentifierOrMarker "," gIdentifierOrMarker "," > uIdentifierOrMarker )? optionalAttributeValuePairs ")" > > I may be wrong, but the ( "," aIdentifierOrMarker "," > gIdentifierOrMarker "," uIdentifierOrMarker )? means that either all > of the terms activity, generation and usage are present or none of > them. In other words, the above definition may need to be altered to > something like: > > derivationExpression ::= "wasDerivedFrom" "(" optionalIdentifier > eIdentifier "," eIdentifier ( "," aIdentifierOrMarker)? ("," > gIdentifierOrMarker ")? (," uIdentifierOrMarker )? > optionalAttributeValuePairs ")" No, we went for a all-or-nothing approach for optionals. > > - Looking at the definition of Revision, Quotation and Primary source, > I am wondering if it would make sense to say something about the > kind of derivation in the derivationExpression, to state that it may > contain an additional optional element that specifies the kind of > derivation. This is already said in prov-dm, and we just wanted to have the minimum necessary to explain the syntax. Thanks, Luc > > Khalid > > On 14 June 2012 12:07, Provenance Working Group Issue Tracker wrote: > > PROV-ISSUE-408 (prov-n-review-for-LC): feedback on PROV-N document (for last call release) [prov-n] > > http://www.w3.org/2011/prov/track/issues/408 > > Raised by: Luc Moreau > On product: prov-n > > > This is the issue to collect feedback on prov-n document. > > Document to review is available from: > > http://dvcs.w3.org/hg/prov/raw-file/default/model/releases/ED-prov-n-20120614/prov-n.html > > Question for reviewers: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2012.06.14 > > Cheers, > Luc >