User:Marek.suchanek/sandbox

Data Stewardship Wizard

Developer(s)	DSW Team
Initial release	October 30, 2018 (2018-10-30)
Stable release	2.2.0 / April 1, 2020 (2020-04-01)
Repository	https://github.com/ds-wizard
Written in	Haskell (back-end) Elm (front-end) Python (workers)
Available in	English
Type	Web application
License	Apache License 2.0
Website	ds-wizard.org

Data Stewardship Wizard (DSW) is an open-source web application that brings together data stewards and researchers to compose data management plans (DMPs) for projects efficiently and in a FAIR manner. Data stewards can easily capture the knowledge, including required project data and decisions in knowledge models that are then turned into per-project questionnaires to be filled by researchers. The questionnaires can be easily edited or customized for specific local regulations and other requirements. Researchers are guided through a questionnaire using recommendations, FAIR metrics indications, and by showing only relevant questions based on the previous answers. With the filled questionnaire, a DMP as a document can be easily generated using a selected template and output format (including PDF, docx, LaTeX, or machine-actionable JSON) and persistently stored directly in the DS Wizard. The benefit lies not only in having a nowadays often obligatory DMP for funders but mainly learning how to handle data correctly, make them FAIR, maintain them well during the project, and curate them long-term.

DSW started as a joint project of Czech and Netherlands nodes of ELIXIR. The content in the form of knowledge model capturing data stewardship expertise has been provided by Duch Techcentre for Life Sciences whereas software design and development is done by a team from Czech Technical University in Prague. The development nowadays continues as DSW participates in several projects. The roadmap is highly influenced by those projects but also feedback from key users such as ELIXIR nodes or SciLifeLab. Source codes can be found in several repositories of ds-wizard organizations at GitHub.

Functionality

• Knowledge Models = A knowledge model captures domain knowledge and sets the structure of questionnaires together with guidance. It can evolve easily over time with published versions according to semantic versioning. A knowledge model can be standalone (built from scratch) or based on another one as customization (so-called fork). When a KM is updated and new version is published, DSW provides easy-to-do migration to the newer version for both customizations and questionnaires.
• Questionnaires = Researchers fill a questionnaire for their project. Each questionnaire is based on some knowledge model that defines the structure and guidance according to the needs. Any time, a DMP document can be generated from the questionnaire using selected template and output format.
• Guidance =
• Document Templates = Document templates used to export questionnaire as DMP are using Jinja2 markup language; therefore, are very versatile and universal. It allows to define templates in practically any textual format including machine-actionable RDF. On the other hand, it is possible to use transformations from HTML to create also PDF, DOCX or LaTeX documents ready for further adjustments and submission to a funder.
• Tags = When creating a questionnaire, a researcher can select only topics relevant for the project or DMP using tags. Data stewards prepare tags in knowledge models related to topics, specific funders, or other categories. Researchers then simply check what is required for their planning and get a filtered questionnaire.
• FAIR metrics = Answers in a knowledge model can affect several metrics in positive or negative way. Currently, there are 6 metrics in DSW - traditional FAIR principles, G for Good DMP Practice and O for Openness. For example, picking sharing results using standard open storage will get better score than storing on laptop in custom format. These metrics are not evaluated as a test but provides indications what options are more suitable than others.
• OpenID = We allow users to login also using external services that are OpenID compliant. That enables organisations to use their own authorization server for login to DSW.

History

// Inception, ELIXIR-CZ portal, Rob Hooft, Barend Mons (link)

Architecture

• Server =
• Client =
• Document Worker =

// Docker, Mongo, RabbitMQ, docs, example

See also

• Data management plan
• FAIR data
• Open science

External links

• Data Stewardship Wizard infoweb
• Data Stewardship Wizard documentation
• Data Stewardship Wizard GitHub organization
• "DEMO" instance of DS Wizard
• "Researchers" instance of DS Wizard