Re: Request for comments - Requirements analysis

Hi Jeni,

> As I said on the call, thank you very much for doing this work, which is very useful. Some comments as I went through it:
> 
> 1. There seem to be a few requirements that are about what should be defined within the data model. Could these be moved to a separate category?
> 

I've created a group "Data Model Requirements" and I've moved there:
- R-HeadingColumns
- R-CellValueMicroSyntax
- R-NonStandardFieldDelimiter
- R-PrimaryKey
- R-ForeignKeyReferences
- R-ExternalDataDefinitionResource
- R-AnnotationAndSupplementaryInfo
- R-AssociationOfCodeValuesWithExternalDefinitions
- R-CsvAsSubsetOfLargerDataset
- R-LinksToExternallyManagedDefinitions
- R-SyntacticTypeDefinition
- R-SemanticTypeDefinition
- R-MissingValueDefinition
- R-URIMapping
- R-UnitMeasureDefinition
- R-GroupingOfMultipleTables

I'm not 100% sure about the completeness and correctness of this division.


> 2. Handling the parsing of CSV-like files into the tabular data model is out of scope for the Working Group (eg R-MultipleHeadingRows) though it’s still useful to capture them. Can we keep them in the document but in a Deferred pile? (These can form evidence that the scope of the WG needs to be expanded when rechartered, should that happen.)
> 
> 3. Handling the provision of CSV files through some kind of access protocol, as in R-RandomAccess is out of scope for the WG. Again, can we put them in the Deferred pile please?
> 

I've moved those two in the "Deferred requirements" section.


> 4. As Davide points out, there are a lot of requirements that feed into R-AnnotationAndSupplementaryInfo, but it’s definitely useful to have the more granular requirements. What I was wondering was whether there were any requirements in R-AnnotationAndSupplementaryInfo that aren’t covered in the more granular requirements, and if so what they were?

I think that in general R-AnnotationAndSupplementaryInfo captures annotations about information like provenance or copyright that are not explicitly covered by the more granular requirements.

Davide

Received on Thursday, 1 May 2014 13:13:02 UTC