Changes for page SDMX 3.0 Standards. Section 1. Framework for SDMX technical standards

Last modified by Helena on 2025/09/10 11:19

From 4.7 to 4.8 From 4.10 to 5.1

From version 4.8

edited by Helena
on 2025/05/14 15:26

Change comment: There is no comment for this version

To version 4.10

edited by Helena
on 2025/05/14 15:28

Change comment: There is no comment for this version

Raw
Rendered

Summary

Page properties (1 modified, 0 added, 0 removed)

Details

Page properties

Content

@@ -154,7 +154,7 @@
  It is important to note that SDMX is primarily focused on the //exchange// and //dissemination// of statistical data and metadata. There may also be many uses for the standard model and formats specified here in the context of internal processing of data that are not concerned with the exchange between organizations and users, however. It is felt that a clear, standard formatting of data and metadata for the purposes of exchange and dissemination can also facilitate internal processing by organizations and users, but this is not the focus of the specification.
--== {{id name="_Toc56233"/}}3.2 SDMX and Process Automation ==
++== 3.2 SDMX and Process Automation ==
  Statistical data and metadata exchanges employ many different automated processes, but some are of more general interest than others. There are some common information technologies that are nearly ubiquitous within information systems today. SDMX aims to provide standards that are most useful for these automated processes and technologies.
@@ -170,7 +170,7 @@
  The SDMX standards specified here are designed to support the requirements of all of these automation processes and technologies.
--== {{id name="_Toc56234"/}}3.3 Statistical Data and Metadata ==
++== 3.3 Statistical Data and Metadata ==
  To avoid confusion about which "data" and "metadata" are the intended content of the SDMX formats specified here, a statement of scope is offered. Statistical "data" are sets of often numeric observations which typically have time associated with them. They are associated with a set of metadata values, representing specific concepts, which act as identifiers and descriptors of the data. These metadata values and concepts can be understood as the named dimensions of a multi-dimensional co-ordinate system, describing what is often called a "cube" of data.
@@ -190,9 +190,9 @@
  [[image:SDMX 3-0-0 SECTION 1 FINAL-1.0_en_a3e7967f.png||height="921" width="629"]]
--**Figure 1: High Level Schematic of Major Artefacts in the SDMX 3.0 Information Model**
++Figure 1: High Level Schematic of Major Artefacts in the SDMX 3.0 Information Model
--== {{id name="_Toc56235"/}}3.4 The SDMX View of Statistical Exchange ==
++== 3.4 The SDMX View of Statistical Exchange ==
  Version 1.0 of ISO/TS 17369 SDMX covered statistical data sets and the metadata related to the structure of these data sets. This scope was useful in supporting the different models of statistical exchange (bilateral exchange, gateway exchange, and data-sharing) but was not by itself sufficient to support them completely. Versions 2.0 and 2.1 provide a much more complete view of statistical exchange, so that an open data-sharing model can be fully supported, and other models of exchange can be more completely automated. In order to produce technical standards that will support this increased scope, the SDMX Information Model provides a broader set of formal objects which describe the actors, processes, and resources within statistical exchanges.
@@ -226,23 +226,16 @@
  * //**Provision Agreement (Metadata Provision Agreement):**// The set of information which describes the way in which data sets and metadata sets are provided by a data/metadata provider. A provision agreement can be constrained in much the same way as a data or metadata flow definition. Thus, a data provider can express the fact that it provides a particular data flow covering a specific set of countries and topics, Importantly, the actual source of registered data or metadata is attached to the provision agreement (in terms of a URL). The term “agreement” is used because this information can be understood as the basis of a “service-level agreement”. In SDMX, however, this is informational metadata to support the technical systems, as opposed to any sort of contractual information (which is outside the scope of a technical specification). In version 3.0, metadata provision agreement and data provision agreement are two separate artefacts.
  * //**Constraint:**// Data and Metadata Constraints describe a subset of a data source or metadata source, and may also provide information about scheduled releases of data. They are associated with data / metadata providers, provision agreements, data flows, metadataflows, data structure definitions and metadata structure definitions.
  * //**Structure Map: **//Structure maps describes a mapping between data structure definitions or dataflows for the purpose of transforming a data set into a different structure. The mapping rules are defined using one or more component maps which each map in turn describes how one or more components from the source data 534 structure definition map to one or more components in that of the target. Represent maps act as lookup tables and specific provision is made for mapping dates and times.
++* //**Representation Map:**// Representation maps describe mappings between source value(s) and target value(s) where the values are restricted to those in a code list, value list or be of a certain type such as integer or string.
++* //**Item Scheme Map:**// An item scheme map describes mapping rules between any item scheme with the exception of code lists and value lists which use representation maps. The version 3.0 information model provides four item scheme maps: organisation scheme map, concept scheme map, category scheme map and reporting taxonomy map. Organisation scheme map and reporting scheme map have been omitted from the information model schematic in Figure 1.
++* //**Reporting Taxonomy: **//A reporting taxonomy allows an organisation to link (possibly in a hierarchical way) a number of cube or data flow definitions which together form a complete “report” of data or metadata. This supports primary reporting which often comprises multiple cubes of heterogeneous data, but may also support other collection and reporting functions. It also supports the specification of publications such as a yearbook, in terms of the data or metadata contained in the publication.
++* //**Process:**// The process class provides a way to model statistical processes as a set of interconnected //process steps.// Although not central to the exchange and dissemination of statistical data and metadata, having a shared description of processing allows for the interoperable exchange and dissemination of reference 556 metadata sets which describe processes-related concepts.
++* //**Hierarchy**//: Describes complex code hierarchies principally for data discovery purposes. The codes themselves are referenced from the code lists in which they are maintained.
++* //**Hierarchy Association**//: A hierarchy association links a hierarchy to something that needs it like a dimension. Furthermore, the linking can be specified in the context of another object such as a dimension in the context of a dataflow. Thus, a dimension in a data structure definition could have different hierarchies depending on the dataflow.
++* //**Transformation Scheme:**// A transformation scheme is a set of Validation and Transformation Language (VTL) transformations aimed at obtaining some meaningful results for the user (e.g., the validation of one or more data sets). The set of transformations is meant to be executed together (in the same run) and may contain any number of transformations in order to produce any number of results. Thus, a transformation scheme can be considered as a VTL ‘program’.
--• //**Representation Map:**// Representation maps describe mappings between source value(s) and target value(s) where the values are restricted to those in a code list, value list or be of a certain type such as integer or string.
++== 3.5 SDMX Registry Services ==
--• //**Item Scheme Map:**// An item scheme map describes mapping rules between any item scheme with the exception of code lists and value lists which use representation maps. The version 3.0 information model provides four item scheme maps: organisation scheme map, concept scheme map, category scheme map and reporting taxonomy map. Organisation scheme map and reporting scheme map have been omitted from the information model schematic in Figure 1.
--
--• //**Reporting Taxonomy: **//A reporting taxonomy allows an organisation to link (possibly in a hierarchical way) a number of cube or data flow definitions which together form a complete “report” of data or metadata. This supports primary reporting which often comprises multiple cubes of heterogeneous data, but may also support other collection and reporting functions. It also supports the specification of publications such as a yearbook, in terms of the data or metadata contained in the publication.
--
--• //**Process:**// The process class provides a way to model statistical processes as a set of interconnected //process steps.// Although not central to the exchange and dissemination of statistical data and metadata, having a shared description of processing allows for the interoperable exchange and dissemination of reference 556 metadata sets which describe processes-related concepts.
--
--• //**Hierarchy**//: Describes complex code hierarchies principally for data discovery purposes. The codes themselves are referenced from the code lists in which they are maintained.
--
--• //**Hierarchy Association**//: A hierarchy association links a hierarchy to something that needs it like a dimension. Furthermore, the linking can be specified in the context of another object such as a dimension in the context of a dataflow. Thus, a dimension in a data structure definition could have different hierarchies depending on the dataflow.
--
--• //**Transformation Scheme:**// A transformation scheme is a set of Validation and Transformation Language (VTL) transformations aimed at obtaining some meaningful results for the user (e.g., the validation of one or more data sets). The set of transformations is meant to be executed together (in the same run) and may contain any number of transformations in order to produce any number of results. Thus, a transformation scheme can be considered as a VTL ‘program’.
--
--== {{id name="_Toc56236"/}}3.5 SDMX Registry Services ==
--
  In order to provide visibility into the large amount of data and metadata which exists within the SDMX model of statistical exchange, it is felt that an architecture based on a set of registry services is potentially useful. A “registry” – as understood in webservices terminology – is an application which maintains and stores metadata for querying, and which can be used by any other application in the network with sufficient access privileges (though note that the mechanism of access control is outside of the scope of the SDMX standard). It can be understood as the index of a distributed database or metadata repository which is made up of all the data provider’s data sets and reference metadata sets within a statistical community, located across the Internet or similar network.
  Note that the SDMX registry services are not concerned with the storage of data or reference metadata. The assumption is that data and reference metadata lives on the sites of its data and metadata providers. The SDMX registry services concern themselves with providing visibility of the data and reference metadata, and information needed to access the data and reference metadata. Thus, a registered data set will have its URL available in the registry, but not the data itself. An application which wishes to access that data would query the registry, perhaps by drilling down via a Category Scheme and Dataflow, for the URL of a registered data source, and then retrieve the data directly from the data provider (using an SDMX REST API query message or other mechanism).
@@ -256,7 +256,7 @@
  * //**Querying: **//The registry services have interfaces for querying the metadata contained in a registry, so that applications and users can discover the existence of data sets and reference metadata sets, structural metadata, the providers/agencies associated with those objects, and the provider agreements which describe how the data and metadata are made available, and how they are categorized.
  * //**Subscription/Notification:**// It is possible to “subscribe” to specific objects in a registry, so that a notification will be sent to all subscribers whenever the registry objects are updated.
--== {{id name="_Toc56237"/}}3.6 RESTful Web services ==
++== 3.6 RESTful Web services ==
  Web services allow computer applications to exchange data directly over the Internet, essentially allowing modular or distributed computing in a more flexible fashion than ever before. In order to allow web services to function, however, many standards are required: for requesting and supplying data; for expressing the enveloping data which is used to package exchanged data; for describing web services to one another, to allow for easy integration into applications that use other web services as data resources.
@@ -270,7 +270,7 @@
  The following conceptual example uses the ‘data’ resource to query a data repository for a series identified by the key ‘M.USD.EUR.SP00.A’ in the EXR (ECB exchange rates) Dataflow: https:~/~/ws-entry-point/data/dataflow/ECB/EXR/1.0.0/M.USD.EUR.SP00.A
--= {{id name="_Toc56238"/}}4 The SDMX Information Model =
++= 4 The SDMX Information Model =
  SDMX provides a way of modelling statistical data, and defines the set of metadata constructs used for this purpose. Because SDMX specifies a number of transmission formats for expressing data and structural metadata, the model is used as a mechanism for guaranteeing that transformation between the different formats is lossless. In this sense, all of the formats are syntax-bound expressions of the common information model.
@@ -288,9 +288,9 @@
  A full UML conceptual design of the information model is set out in Section 2 of the Technical Specifications.
--= {{id name="_Toc56239"/}}5 The SDMX Transmission Formats =
++= 5 The SDMX Transmission Formats =
--== {{id name="_Toc56240"/}}5.1 SDMX-ML ==
++== 5.1 SDMX-ML ==
  SDMX-ML is the XML transmission format specification for exchanging structural metadata, data and reference metadata, and interacting with SDMX registry services. It is designed as a general-purpose format for all automation and data / metadata exchange tasks, and provides the most complete coverage.
@@ -302,12 +302,10 @@
  Many XML tools and technologies have expectations about the functions performed by an XML schema, one of which is a very direct relationship between the XML constructs described in the XML schema and the tagged data in the XML instance. Strong data typing is also considered normal, supporting full validation of the tagged data. These message types are designed to support validation and other expected XML schema functions.
 . //Generic Metadata~:// For the exchange of reference metadata sets. ‘Generic’ means the XML elements and XML attributes are the same regardless of the metadata set.
--1. //Registry~:// All of the possible interactions with the SDMX registry services are supported using SDMX-ML interfaces and REST API calls. Submission of structural metadata content, data / metadata registrations and subscriptions is performed by a synchronous exchange of documents – a “request” message answered by a
++1. //Registry~:// All of the possible interactions with the SDMX registry services are supported using SDMX-ML interfaces and REST API calls. Submission of structural metadata content, data / metadata registrations and subscriptions is performed by a synchronous exchange of documents – a “request” message answered by a “response” message.
--“response” message.
++== 5.2 SDMX-JSON ==
--== {{id name="_Toc56241"/}}5.2 SDMX-JSON ==
--
  SDMX-JSON is the JSON transmission format specification for exchanging structural metadata, data and reference metadata. It provides an alternative to SDMX-ML and is most suited to applications like web data dissemination.
  SDMX-JSON messages serve the same function as those of the XML formats but have a different structure. For data, an important distinction is that they carry both component codes and labels which provides all the information needed to display the content in a single JSON response. The XML Structure-specific Data format by contrast carries only code IDs thus requiring applications obtain and hold structural metadata about the data set in order to display the content in human-readable form.
@@ -320,7 +320,7 @@
 . //Data: //For the exchange of data. Unlike SDMX-ML, the structure of a SDMX-JSON data message is not specific to the DSDs of the data sets so schema validation will not check for compliance of the data with the DSDs.
 . //Metadata//: For the exchange of reference metadata sets.
--== {{id name="_Toc56242"/}}5.3 SDMX-CSV ==
++== 5.3 SDMX-CSV ==
  SDMX-CSV is the CSV transmission format specification for exchanging data and reference metadata only.
@@ -331,7 +331,7 @@
 . //Data//: For the exchange of data. Like SDMX-JSON, SDMX-CSV can include both code IDs and labels which is helpful when using the data to create human readable charts and dashboards.
 . //Metadata//: For the exchange of reference metadata sets.
--== {{id name="_Toc56243"/}}5.4 Formats and Messages Deprecated in Version 3.0 ==
++== 5.4 Formats and Messages Deprecated in Version 3.0 ==
  The following formats and messages have been deprecated in version 3.0 to simplify, modernise and rationalise the standard.
@@ -348,35 +348,35 @@
  * SDMX-ML Query messages
  * SDMX-ML Submit Structure Request messages
--= {{id name="_Toc56244"/}}6 Dependencies on SDMX content-oriented guidelines =
++= 6 Dependencies on SDMX content-oriented guidelines =
  The technical standards proposed here are designed so that they can be used in conjunction with other SDMX guidelines which are more closely tied to the content and semantics of statistical data exchange. The SDMX Information Model works equally well with any statistical concept, but to encourage interoperability, it is also necessary to standardize and harmonize the use of specific concepts and terminology. To achieve this goal, SDMX creates and maintains guidelines for cross-domain concepts, terminology, and structural definitions. There are three major parts to this effort.
--== {{id name="_Toc56245"/}}6.1 Cross-Domain Concepts ==
++== 6.1 Cross-Domain Concepts ==
  The SDMX Cross-Domain Concepts is a content guideline concerning concepts which are used across statistical domains. This list is expected to grow and to be subject to revision as SDMX is used in a growing number of domains. The use of the SDMX Cross-Domain Concepts, where appropriate, provides a framework to further promote interoperability among organisations using the technical standards presented here. The harmonization of statistical concepts includes not only the definitions of the concepts, and their names, but also, where appropriate, their representation with standard code lists, and the role they play within data structure definitions and metadata structure definitions.
  The intent of this guideline is two-fold: to provide a core set of concepts which can be used to structure statistical data and metadata, to promote interoperability between systems (“structural metadata”, as described above); and to promote the exchange of metadata more widely, with a set of harmonized concept names and definitions for other types of metadata (“reference metadata”, as defined above.)
--== {{id name="_Toc56246"/}}6.2 Metadata Common Vocabulary ==
++== 6.2 Metadata Common Vocabulary ==
  The Metadata Common Vocabulary is an SDMX guideline which provides definition of terms to be used for the comparison and mapping of terminology found in data structure definitions and in other aspects of statistical metadata management. Essentially, it provides ISOcompliant definitions for a wide range of statistical terms, which may be used directly, or against which other terminology systems may be mapped. This set of terms is inclusive of the terminology used within the SDMX Technical Standards.
  The MCV provides definitions for terms on which the SDMX Cross-Domain Metadata Concepts work is built.
--== {{id name="_Toc56247"/}}6.3 Statistical Subject-Matter Domains ==
++== 6.3 Statistical Subject-Matter Domains ==
  The Statistical Subject-Matter Domains is a listing of the breadth of statistical information for the purposes of organizing widespread statistical exchange and categorization. It acts as a standard scheme against which the categorization schemes of various counterparties can be mapped, to facilitate interoperable data and metadata exchange. It serves another useful purpose, however, which is to allow an organization of corresponding “domain groups”, each of which could define standard data structure definitions, concepts, etc. within their domains. Such groups already exist within the international community. SDMX would use the Statistical Subject-Matter Domains list to facilitate the efforts of these groups to develop the kinds of content standards which could support the interoperation of SDMX-conformant technical systems within and across statistical domains. The organisation of the content of such schemes is supported in SDMX as a Category Scheme.
  SDMX Statistical Subject-Matter Domains will be listed and maintained by the SDMX Initiative and will be subject to adjustment.
--== {{id name="_Toc56248"/}}6.4 SDMX Concept Roles ==
++== 6.4 SDMX Concept Roles ==
  These guidelines define the standard set of SDMX Concept Roles and their use. This set of standard SDMX Concepts are implemented as a cross-domain Concept Scheme that defines the set of concept roles and gives examples on concept role implementation in SDMX 2.0, 2.1 and 3.0. A concept role gives a particular context to a concept for easy and systematic interpretation by machine processing and visualization tools. For example, the concepts REPORTING_AREA and COUNTERPART_AREA are different concepts but they are both geographical characteristics, therefore they can be associated with the same concept role ID: "GEO". This allows visualization systems to interpret these concepts as geographical data in order to generate maps. The implementation of concept roles is different in versions 2.0 and 2.1/3.0 of the SDMX technical standard. Specifically for SDMX 3.0, this set of roles is considered a normative list that must be interpreted in the same way by all organisations.
  Additional roles may be provided via the standard roles’ mechanism in SDMX 3.0, i.e., via Concept Schemes; the semantics of these roles have to be agreed bilateraly in data exchanges. The Concept Roles are available as an SDMX Concept Scheme on the SDMX Global Registry.
--= {{id name="_Toc56249"/}}7 Validation and Transformation Language =
++= 7 Validation and Transformation Language =
  For many years the SDMX initiative has been fostering and supporting the development of a standard calculation language, called Validation and Transformation Language (VTL). A blueprint for defining calculations was already described in the original SDMX 2.1 specifications (package 13 of the Information Model - “Transformations and Expressions”). It was just a basic framework that required further developments to became operational in order to achieve a calculation language able to manipulate SDMX artefacts.