Data Governance Policy

Data Scope and Criteria

‍

Type of data which may be shared on DEEP:

• Qualitative (e.g., reports, articles, policy documents, needs assessments and response data) and quantitative data (e.g., dataset on impacted and vulnerable people and communities) in format supported by DEEP
• Any type of data that supports a project’s objective

Criteria of data shared on DEEP:

• Public and restricted sources (leads) may not contain any personal data. Personal data is information, in any form, that relates to an identified or identifiable natural person.
• Public and restricted sources (leads) may not contain any sensitive non-personal data.This includes information which, while not relating to an identified or identifiable natural person, may, by reason of its sensitive context, put certain individuals or groups of individuals at risk of harm.
• Metadata of public and restricted sources are made available for all DEEP users as public view.
• Confidential sources (leads) may contain personal and sensitive data if in line with project objectives and treated with utmost confidentiality such as restricted user access and/or set-up of the project for private access only. Metadata for a private project is excluded for public view.
• Should copyrights apply to asource, appropriate restriction level should be applied and information there of included in the source metadata.

‍

Data responsibility and sharing

‍

Responsible publisher. The Project Administrator is the responsible publisher for his/her project and must ensure that collaborators ofthe project follow DEEP’s data sharing protocols. The Project Administrator needs to sign a Data sharing agreement with indication of level of openness of data.

Users may use DEEP to share data from other sources with appropriate metadata including any applicable copyright,restrictions or license to guide any onward sharing.

‍

Restriction levels. There are three levels of restriction impacting data sharing on DEEP:

• Public: Data can be shared with any user in DEEP and exported to anyone without restrictions. The data is unlikely to cause any harm to affected population orother individuals and is publicly available. The data can be shared across the platform for research purposes (for example development of NLP), available in joint repository and platform cross searches.
• Restricted: Data is accessible only to users collaborating on the project where the data is uploaded on DEEP. Data might be sensitive or under copyright. The metadata of a dataset is available and the [project owner] can decide whether or not to grant access to the full dataset when requested by a user. The data is not accessible in platform cross searches, joint repository or for other purposes. Other users will only be able to consult the metadata.
• Confidential: Data is considered confidential and could cause harm to affected population or other individuals. The metadata of a dataset is available and the [project owner] can decide whether or not to grant access to the dataset when requested by a user, but should be shared only with caution and due procedure.The data is not accessible in platform cross searches, joint repository or for other purposes. Other users will only be able to consult the metadata.

A project can be set-up as a private project which exclude all information in the project including metadata from the public view. Contact support@thedeep.io, to ask to set up as a private project.

After exporting data from DEEP, users must follow any restrictions applied when using and further sharing the data. Exports including confidential data will be marked and restriction metadata included inbibliography.

If a user notices personal or sensitivedata shared publicly through DEEP they should contact support@thedeep.io immediately to request that the data be removed. The DEEP team will take action, removing public access to that data and contact the project administrator.

‍

Data Management

‍

Open source. DEEP is an open-source data management system licensed under the AGPL-3.0 license under the Open Source Initiative and managed by Data Friendly Space (DFS), a United States non-profit (501 c3) based organization.

Server storage. Data that is uploaded to DEEP is stored by DFS on servers provided by Amazon Web Services and long term backups are stored on N2WS. Data is encrypted in transit and at rest. Data is hosted and processed in the United States.

• DFS will never alter any data uploaded to DEEP without permission from the Project Administrator.
• Data shared through DEEP will never be shared further by DFS without prior permission from Project Administrator either for research and/or training dataset for NLP.

Deletion of data. Deleted data cannot be retrieved by users. Deleted sources are not currently purged from backups. Metadata continues to exist in backups of the database indefinitely.

‍

Generic Disclamer of Liability

‍

The users are responsible for the data they share on DEEP. DEEP, its governing members or DFS assumes no liability whatsoever for data shared on DEEP. Sharing data through DEEP does not imply the transfer of any rights over this data to DEEP. DEEP, its governing members or DFS disclaim all warranties, whether expressed or implied.

‍

Data and information on DEEP do not imply the expression or endorsement of any opinion on the part of DEEP, DFS, or its governing members. This includes opinions concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries.

‍