Toggle menu

Open data policy

Dutch version

Why open data?

In its mission statement, the Research Institute for Nature and Forest (INBO) declares to support decision makers in the development of a sustainable society by providing insight in nature and forest through qualitative research. This mission is expressed as set of strategic goals to support governmental initiatives, partners, and other policy makers, and to implement a European biodiversity strategy.

The operational functioning of the institute aims to accomplish these strategic goals by executing scientific research and monitoring. The outcome is communicated in scientific publications, reports and advisory documents, which support decision makers in their choices on management and policy. However, both internal and external stakeholders do not only benefit from the final decision, report or scientific paper, but from the data on which these are based as well. Opening these data does not only contribute to the transparency of the INBO, but also provides an important social, scientific and even commercial added value. Providing these data to the general public in a documented and free to use manner - in other words as open data - is an essential element in the implementation of our strategic goals. This is why the INBO is an open data institute since 2015, whereby open is the norm.

Open data policy

An open data institute requires a clear open data policy. Our open data policy consists of the following 10 guidelines:

1. All scientific data for which the (EV) INBO is (co-)owner are covered by this policy and can be published as open data, excluding legal exceptions.

All scientific data (co)owned by the INBO are covered by this policy. External data for which the INBO has access rights, but is not the (co-)owner, are only covered by this policy when publication as open data is allowed under contract or when the owner has given his written permission to do so. For new projects, contracts should stipulate the INBO has (co-)ownership of the resulting data and has the right to publish those under the conditions of this policy.
Data publication is subject to the legal exceptions regarding the divulgation of environmental information (decree of 26 March 2004, section 15), such as the protection of the environment, personal information, and confidential information. These exceptions are applied to specific attributes of a dataset (e.g. geospatial coordinates), never the dataset as a whole, preferably by anonymising or reducing precision, and are documented in the metadata.

2. Data are eligible for publication at the earliest 12 months after collecting.

This embargo is effective on a record level, i.e. for long-term projects a subset of the growing dataset can be published earlier. The eligibility for the publication does not imply the data will be published, as this depends on prioritization. Nevertheless, individual researchers or projects (e.g. LifeWatch) can opt to open the data earlier.

3. Raw data will be published.

Collected data in a usable form (e.g. species observations or measurements) are raw data. These data can potentially be used for multiple research questions or purposes. Raw data are eligible for publication if 1) there is a specific request for these and/or 2) there is an appropriate infrastructure to disclose these, such as GBIF for observations and taxonomic data, GenBank or similar for genetic data, and GeoPunt for important GIS layers. The publication of raw data is prioritized and planned on a yearly basis.

4. Result data associated with a scientific article, report, or advice will be released with the publication.

Associated result data refers to data which are used and/or discussed in a publication. In contrast to raw data, these data typically went through a transformation process (e.g. aggregations, statistical analyses). The associated data will be released in combination with the publication of the scientific paper, report or advisory document, even when the embargo (guideline 2) is still active. The data are published using a data repository and are referenced in the published work. When suitable, one can also reference the already published raw data.

5. Data are published as open data, as defined by the Open Definition, under the Creative Commons Zero waiver.

All data are published according to the Open Definition, which declares that anyone can freely access, distribute and (re)use the data. Data are published under the Creative Commons Zero (CC0) waiver, whereby the owner waives all intellectual rights and dedicates the data to the public domain. This waiver is legally the most appropriate choice for the type of data managed by the INBO.

6. Published datasets reference the INBO norms for data use.

Since copyright does not apply to factual data(*), one can not enforce user requirements on this type of public data. The INBO does however express its wishes regarding citation and responsible data use - especially in a scientific context - as a set of norms. These norms are referenced in all published datasets.

7. Published data are sufficiently documented with metadata.

Metadata describe a dataset and mention among other things contact information, methodology, and taxonomic, temporal and geographical scope. This information helps the user to better understand the data. All published data will be documented with proper metadata. The format and extent of these metadata are subject to the requirements of the platform (e.g. GBIF, GenBank, Dryad) used to publish the dataset.

8. All upcoming research projects require a data management plan (DMP).

A data management plan (DMP) describes how research data will be managed, documented and shared. A DMP should be drafted at the start of a project. The format and content of the DMP depend on the characteristics of the project and requirements of the project partners and/or funders.

9. Researchers apply the open data policy.

Following the open data policy is included as a general goal in the yearly planning and evaluation of researchers.

10. The INBO Information & Data Center supports researchers in the application of the open data policy.

The INBO Information & Data Center (IDC) implements the open data policy by 1) providing the required support and tools to effectively publish data, 2) facilitating the writing of data papers, and 3) providing support in the standardisation of both data and metadata using international standards.

Conclusion

Open data is the norm for all public institutions in Flanders and Belgium. The European commission also subscribes the importance of open data, both in the context of administrative information as to support research projects. In the scientific community, publishing research data as open data fits in the larger transition to open access and open source. The INBO hopes its open data policy will contribute towards more transparent and accessible research in Flanders concerning nature and biodiversity.

Preferred citation

This page is published under a Creative Commons Attribution license. You can cite it as: 

Desmet P, Brosens D, Du Seuil D, Goossens B, Van Hoey S (2014) Open data policy of the INBO. Research Institute for Nature and Forest (INBO). http://www.inbo.be/en/open-data-policy

 

Related

 

(*) Data are subject to three types of intellectual rights: copyright, database right and sui generis database right. For factual data the first two do not apply, since data is not a creation of the human mind and publication is done in a standardized and/or open format. The sui generis database right also applies in specific cases only: it only applies in some (mainly European) jurisdictions, when the data owner has made a substantial investment in the collection of the data and when the user plans to use a substantial part of the data. By using a CC0 waiver, the owner waives these (often non-existing or reduced) intellectual rights where legally possible or - when legally waving those (e.g. moral) rights is not possible - declares not to use those. Doing so effectively places the data in the public domain, allowing anyone to freely copy, use, modify and distribute the data.