2.2 – Data schemas for aid information

2.2.1 – Standard and non-standard schemas

0 A data schema structures and describes information of a certain kind. Schemas may be non-standard, specific to a particular application or purpose, or they may adhere to existing standards.

0 For example, Jessica may decide to create a catalogue of her books. She might devise her own way of doing this using database software – and make her own decisions about how to catalogue different types of publications, how to record different pieces of information, what information to include and so on. Alternately, she might create her catalogue in adherence to the Machine Readable Cataloging (MARC) standards – which specify how she should catalogue her books, and give a large range of standard fields which she can use as well as details about how the information should be formatted. In the latter case, she will be able to import bibliographic records adhering to the MARC standards, and share her records with others that use MARC. In the former case, the catalogue will most likely not be interoperable with other catalogues – unless perhaps work is done to translate the records in her catalogue to a standard format.

0 Adhering to a standard data schema is important to ensure that datasets can be compared and combined together. Without adhering to a standard, significant work may have to be done to harmonise data from different sources.

2.2.2 – Existing schemas and standards

0 There are many data schemas in different domains that may be of interest to those publishing aid information.

0 In the late 1990s, the International Development Markup Language (IDML) was created to assist with the exchange of international development related information. The Development Gateway became interested in the project in 2001, and are currently developing it ¹.

0 In relation to statistical data, there is the Statistical Data and Metadata eXchange (SDMX) standard, which is supported by many national and international organisation, including the United Nations, the OECD, Eurostat, World Bank, the IMF, and many others.

0 In relation to financial and business information, schemas include the Financial Information eXchange (FIX), the Financial products Markup Language (FpML) and the eXtensible Business Reporting Language (XBRL).

0 In addition to domain specific schemas, there are also cross domain initiatives, such as the Dublin Core standard, which develops metadata standards to facilitate sharing information.

2.2.3 – Which schemas?

0 We propose that further work should be done towards a data schema for the exchange of information related to international development. Such work should take into account existing developments, and collaborate with relevant parties to ensure interoperability with existing schemas and compliance with existing standards.