Data has quickly become the “New Oil” in current digital economy. In the past, data were considered to be part of an application program through which they are created/input, processed and results are computed/derived. Today, data has taken the centre stage, and many innovative applications and programs are now created to process, analyse and derive useful insights from the data. However, many a time, datasets are created or produced and made available on an “as-is” or ad-hoc basis. To ready these datasets so that they could be meaningfully consumed by the applications and also used to facilitate interoperability across multiple systems, concerted effort and care needs to be put in to collect, pre-process, format, organise and package these datasets according to specified standards, guidelines and codes of practice.
TR 33 : 2013 Data as a service (DaaS) application programming interface (API) design and implementation
Data as a Service (hereafter referred to as DaaS) allows organisations to share, commoditise or monetise data as an on-demand service. Users may then, directly or through applications, analyse, process, combine and present these data as information or as part of other value-added services. This Technical Reference (TR) recommends a set of best practices in the implementation of data APIs. It also includes an example of how an API for a DaaS site shall be documented as well as an implementation checklist. These recommendations and supporting documentation are designed to give developers or independent software vendors (ISVs) as much agility as possible when rolling out new services, and ensuring that these services can easily be maintained even as changes are made to source datasets.
TR 41 : 2015 Data quality metrics
This set of guidelines recommends a baseline set of data quality metrics which is industry domain agnostic, for adoption by data providers. In addition to describing the methodology for deriving this set of metrics, tools for relaying metrics to end-users are also considered. Having a common set of metrics allows users to more easily compare the quality of different datasets, and match their expectations against available datasets. This set of guidelines articulates and defines a common set of domain agnostic data quality metrics for structured and machine-readable datasets; data quality metrics for unstructured datasets are out of the scope.
TR 55 : 2016 Data versioning
This Technical Reference recommends a minimum set of coherent international or industry standards for interface interoperability of information and services that support a variety of applications across multiple industries and are suitable for deployment on a nation-wide scale.
Where to Buy
The Singapore data standards can be purchased from the Singapore Standards eShop.