Use-case,includes format transformation,target output provided,column names mapped to properties with QNames,cell values mapped to resources with QNames,"""variables"" present in cells",complex structure in output,conditional rules / flow control,string manipulation
Use Case #1 - Digital preservation of government records,"yes - uses combination of XSLT, Java and Scala to create both RDF/XML data and XHTML documentation",no,yes - but unclear if an external vocabulary is used,yes - reference to other Records in the CSV to create a hierarchy,no example provided,unknown,unknown,unknown
Use Case #2 - Publication of National Statistics,no,n/a,no - but the use case notes the need to define explicit semantics for columns; but this could be done in the CSV schema,yes - examples include topic categories (e.g. T061A) and geographic area codes,n/a,n/a,n/a,n/a
Use Case #3 - Creation of consolidated global land surface temperature climate databank,no - but the CSV content does need to be packaged in such a way that it can be readily consumed by custom (scientific) software; currently Fortran95 but latterly IDL and Python ,n/a,no - but the need for unambiguous semantics for columns is present,yes - examples include mapping to resources in other CSV files; such as station identifiers (e.g. REC41011874) and place names (e.g. ENTEBBE_AIRPO) ,"n/a - but fixed formats are used in some text fields suggesting that ""variables"" might be extracted independently",n/a,n/a,n/a
Use Case #4 - Publication of public sector roles and salaries,yes - RDF is used as an interim format to support data visualisation widget,yes - RDF/XML,"yes - largely a local ""organogram"" schema, but also uses the RDF Data Cube vocab (qb)",yes - also see Jeni's experiments for UC4 [https://github.com/w3c/csvw/blob/gh-pages/examples/tests/scenarios/uc-4/attempts/attempt-2/README.md],no,"no - but the RDF/XML output does embed properties within objects identified by URI; e.g.
Operational Delivery
",no,no
Use Case #5 - Publication of property transaction data,yes - conversion to RDF for ingestion into triplestore and exposure via SPARQL endpoint,"no - the document resulting from the transformation is not available, but the RDF can be accessed via a public SPARQL endpoint",yes,"yes - examples include unique transaction identifiers, predefined category codes, ""addressable object codes"", local authorities and (probably) geographical area codes",(probably) no,(probably) no,(probably) no,(probably) no
Use Case #6 - Journal Article Solr Search Results ,no - but the search results are provided in XML and JSON alongside CSV,yes - both XML and JSON,no - column titles map straight to XML element names / JSON object keys,"no - but the ""id"" field is a candidate for this",no - but delimited lists of authors are presented as arrays,no,no,no
Use Case #7 - Reliability Analyzes of Police Open Data,no,n/a,yes - enabling axiomatic mappings to be established between datasets with different category criteria,yes - examples include [police] force and neighbourhood code,no,n/a,n/a,n/a
Use Case #8 - Analyzing Scientific Spreadsheets,no,n/a,,,no,"n/a - but there is a need to express complex things such as unit of measurement applied to the observed value (although the RDF Data Cube vocab allows this to be done at the ""slice"" level rather than each value)",n/a,n/a
Use Case #9 - Chemical Imaging,no,n/a,,,"no - but the ""info"" field looks like structured text",n/a - ditto above about unit of measurement,n/a,n/a
Use Case #10 - OpenSpending Data,no,n/a,,,no example provided,n/a - ditto above about unit of measurement,n/a,n/a
Use Case #11 - City of Palo Alto Tree Data,no - this is about Google Fusion tables and visualisation on a map,n/a - although the fusion tables and maps are public,probably,"yes - e.g. ""tree ID""","yes - includes KML markup; e.g.
-122.156485,37.440963
... Also note ""comments"" column is a semi-colon delimited list",n/a,n/a,n/a
Use Case #12 - Chemical Structures,no,n/a,,,no,no - but this use case introduces multiple tables in a single CSV file,n/a,n/a
Use Case #13 - Representing Entities and Facts Extracted From Text,yes - RDF,no,yes,yes - referencing other entities in related CSV files and externally defined properties e.g. per:parent,"yes(ish) - provenance statements for an inferred statement are expressed on a single row using a structured, repeating form - but this could be easily refactored as illustrated in the use case",no,no,no
Use Case #14 - Displaying Locations of Care Homes on a Map,yes - GeoJSON ... Although the use case talks about providing a native API to CSV to avoid the intermediate GeoJSON encoding,no,"yes - e.g. lat, lon etc.",no,no,no,no,no
Use Case #15 - Intelligently Previewing CSV files,yes - JSON,no,no,no,no,no,no,no
Use Case #16 - Tabular Representations of NetCDF data Using CDL Syntax,no,n/a,,,no ,no,no,no
Use Case #17 - Canonical mapping of CSV,yes - JSON,no example given,no example given,no example given,no example given,no,no,no
Use Case #18 - Supporting Semantic-based Recommendations,no,n/a,,,,,,
Use Case #19 - Supporting Right to Left (RTL) Directionality,no,n/a,,,no,n/a,n/a,n/a
Use Case #20 - Integrating components with the TIBCO Spotfire platform using tabular data,no,n/a,,,yes,n/a,n/a,n/a
Use Case #21 - Publication of Biodiversity Information from GBIF using the Darwin Core Archive Standard,yes - GeoJSON,yes ... But it's a contrived example to display the info on a map using Leaflet,"yes - defined in ""DarwinCore""",yes - external taxonomies,no,"not apparantly - but this use case introduces the need to include ""default"" property-value pairs if a given cell is empty",no,no
Use Case #22 - Making sense of other people's data,no,n/a,yes,yes,no,n/a,n/a,n/a
Use Case #23 - Collating humanitarian information for crisis response,no - but conversion to RDF is a longer-term goal,no,"yes - to ""HXL data dictionary""",yes,no,no - but this use case introduces multiple columns (for the same property) to carry repeated values of that property,n/a,n/a
Use Case #24 - Expressing a hierarchy within occupational listings,yes - RDF (SKOS),yes,yes - SKOS,yes - references between terms in the two occupational listings,yes - the SOC code is structured,yes - but the SKOS broader relationships are transitive so can be inferred ... And could be created with a SPARQL CONSTRUCT statement post-fact,yes,no
,,,,,,,,
other examples & experiments,,,,,,,,
simple-weather-observation (jeremy),,,,,,,,yes - to create a URI from a textual (datetime) field
look-inside-art (danbri),,,,,,,,
simple-templates-jquery (ivan),,,,,,,,
,,,,,,,,