Map data format and quality

Please note that the Act on Information Management in Public Administration (Finlex)Opens in a new window. obliges the authorities to provide data in a commonly used machine-readable format if the data can be converted directly from its original form into a machine-readable format.

The Act specifically states that datasets defined as high-value by the EU must be available in a machine-readable format. Read more on about high-value datasets on the Pay attention to laws and provisions page of this guide.

Metadata is information about data on data. It describes and defines the actual data. Metadata can also be called descriptive data or ancillary information.

Information that describes the context, content or structure of a dataset and guides and documents its processing and management.

- Definition of metadata in the Tietotermit glossary (2018)

See the Glossary in the Interoperability Platform for the definition (sanastot.suomi.fi).Opens in a new window.

A data model consists of metadata that describe the structure of the data. The data model is a kind of “reading instruction” for understanding the relationships between the data contained in the dataset.

[A data model is] a description of the relationships between data and information.

- TEPA term bank

Please note that in this guide, a data model refers in particular to a logical data model, not to physical data models or concept models.

Definition of the term logical data model in the Interoperability Platform's Glossary.Opens in a new window.

If the metadata and data model of the dataset are poorly designed,

• utilisation of the data in an information systems is challenging
• there is a risk that the meaning of the data contained in the dataset will not be preserved and changes will not be traceable
• it is unlikely to be easy to find the data.

For example, metadata may need to be added to the dataset. It may also be necessary to develop the data's data model. If the dataset is extensive, the workload may be large.

Opening access to data as open data requires that the data is shared in a machine-readable, structured format, i.e. either as a file or through APIs.

The

• closer the data's reference date is to the current date, the more up-to-date the data is
• more harmonised and non-conflicting the datasets, the more consistent the data is
• the better the dataset contains the desired data, the more comprehensive the data is
• the better data in the dataset corresponds to reality, the more accurate the data is.

The

• better the origin of the data is known, the more traceable the data is
• the more comprehensively data has metadata, the more comprehensible the data is
• the better the dataset and its characteristics comply with known standards, practices and regulations, the more in line with the recommendations the data is.

The

• more structured the format of the dataset, the easier the data is to utilise
• easier the dataset is to access at the indicated times, the more timely the data is.

The usability of the data is also affected by how the access rights to the dataset have been defined and the purposes for which the dataset can be utilised.