Sdmx2 context
Download as PPTX, PDF0 likes370 views
This document discusses information standards for statistical data exchange. It describes key standards like SDMX and DDI, which foster interoperability among statistical organizations. SDMX focuses on time series data exchange while DDI specifies metadata for social science data. The document contrasts schema-based standards, which have separate schema and data layers, from artefact-based standards, which add a domain-specific artefact layer. It provides examples like SDMX and Dublin Core to illustrate the differences. Finally, the document discusses requirements for information standards to be implementation friendly, developer friendly and user friendly.
Sdmx2 context
- 1. { SDMX 2 Context UN/FAO, Erik van Ingen Abuja, Nigeria, 13-14 May 2015
- 2. all the people could find – use - load – display - interpret - interchange – analyze all statistical data Imagine
- 3. Harmonization Standardization Interoperability Better data at lower cost Business Case
- 4. SDMX - Statistical Data and Metadata eXchange - Foster standards for the exchange of statistical information. DDI - Data Documentation Initiative - A metadata specification for the social and behavioral sciences DarwinCore, OGC and ISO Observations and Measurements , Odata, etc. Information Standards
- 5. Standard setting bodies standard body SDMX BIS - ECB - EUROSTAT - IMF - OECD - UN - World Bank DDI Widespread community DarwinCore Widespread community OGC Number of Active Members: 509 Observations and Measurements ISO
- 6. Observation data (also referred to as raw data or operational data ) is a measure taken at a certain moment in time (DDI) Statistical data (timeseries) is often processed and aggregated observation data (SDMX) Observation data & Statistical data Sample # pH 1 6.39 2 6.42 3 6.35 4 6.85
- 7. GSBPM Generic Statistical Business Process Model from United Nations Economic Commission for Europe (UNECE)
- 8. GSBPM The GSBPM is a reference model for the ‘statistical data process’ The GSBPM can help to put information standards within a context
- 9. workflow
- 11. Schema based has 2 layers: 1. schema (domain specific) 2. data Artefact based has 3 layers: 1. schema 2. artefact (domain specific) 3. data Schema or artefact based
- 12. For the techies in the audience: A schema is often an XML schema An artefact of often an XML document The data is often an XML document, also called a message Note: XML has a successor called JSON, which is more lightweight, less verbose and web friendly Schema or artefact based
- 13. An information standard can be schema or artefact based. Schema or artefact refers to the type of information standard. Or: it refers to the style or approach taken when the information standard was developed. Schema or artefact based
- 14. Schema based Schema (metadata, domain specific*1) ) Data (message) depends directly on the *1) In this presentation we boldly consider all schema based standards as domain specific. In reality this is not always the case, because a domain generic schema based standard can have domain specific extensions.
- 15. Artefact based Schema (metadata) Data (message) depends on the Artefact (domain specific metadata) Depends on the
- 16. Example Schema based UN/CEFACT FLUX Schema VMS data depends directly on the
- 17. Example Artefact based SDMX Schema SDMX catch timeseries depends on the Catch DSD (data structure definition) Depends on the
- 18. characteristics pro con examples schema based often domain specific 2 layers Single Maintenance Model less freedom pragmatic easier to implement central led harmonization every domain has its own standard DublinCore artefact based not domain specific 3 layers Delegated Maintenance Model more freedom ‘organic’ harmonization higher complexity DDI SDMX
- 19. How is a codelist represented? schema based In the schema, often as a enumeration. This is however not practical for long lists. When it comes to long lists, a merely schema based standard uses an artefact approach, as XML, CSV or Text file. artefact based As an artefact.
- 20. MDM Capacity for the management of Reference data Code lists Classifications Mappings Harmonization and its related workflow Master Data Management (MDM)
- 21. The business layer defines the metadata of the information (schema or artefact based). “What is in the truck?“ The transport layer defines how the data and metadata will be transmitted. “What are the truck and the road?” Business & Transport layer
- 22. implementation friendly developer friendly user friendly Requirements (related to information standards)
- 23. thank you