Apply for class 03 now!

How OpenRedact Found a New Home on openCode

  • Knowledge Sharing
Foto von aufeinandergestapelten, historischen Reisekoffern, die im Deutschen Technikmuseum in Berlin ausgestellt sind.

In open-source projects that emerge as part of funding programmes like the Prototype Fund, crucial questions often arise even before the funding period ends: What are the next steps for the project? Who will continue to contribute, and in what way? And who will take on key roles to ensure the project’s continued existence?

In the case of OpenRedact, the response was unusual: the software for anonymising personal data in documents was identified by the Datenlabor (data laboratory) of the German Federal Ministry of Research, Technology and Space (BMFTR) as a suitable tool for its own use. The Ministry subsequently took on a leading role in the project’s further development in order to put OpenRedact into use. This involved both improving security and migrating the project to the openCode platform, the federal government’s GitLab instance, where open-source projects relevant to public administration are published and developed.

In this blog post, we take a look at the collaborative handover process, the adjustments made, and what lessons can be learned from this to ensure the long-term sustainability of open-source projects in public administration.

After funding: Public sector organisations as active participants in FOSS projects

Funding programmes, which only run for a limited time, such as the Prototype Fund are designed to turn ideas into working prototypes. While this is often successful, the equally important question of how projects can be sustained and maintained comes up. Without ongoing maintenance, software projects quickly become unusable or pose risks to users.

In many Free and Open Source Software (FOSS) projects, major users are also among the most important contributors. These are often companies, within which roles, funding models and contribution processes have become established over the years. One example of this is the practice of employing maintainers. Public sector organisations have taken on this role far less frequently so far. However, they too can benefit from investing in open-source projects relevant to their tasks, developing them further and maintaining them in collaboration with others. Taking an active role in FOSS projects also gives public sector organisations the opportunity to promote technologies that benefit the common good rather than just economic profitability: such as transparency, digital independence from large tech companies and democratic processes. The takeover of OpenRedact by the BMFTR's Datenlabor illustrates what such a role might entail. In this case, the public sector acts not merely as a user, but invests resources and takes on responsibility for the long-term stabilisation and further development of a FOSS project.

What is OpenRedact and who was it developed for?

OpenRedact was initially developed in 2020 as part of the seventh funding round of the Prototype Fund, and it addressed a structural problem: public bodies, such as ministries or courts, are required to publish documents while at the same time protecting personal data. This is an important transparency measure, but one that can be costly. The project team developed an application that automatically scans documents for personal information and assists with its anonymisation. The software recognises various types of personal data, including names, addresses, dates of birth, telephone numbers. It was supplemented by a legally sound anonymisation guide, which offers users practical guidance on redacting personal information.

In the years following the funding, public sector organisations and their IT service providers had repeatedly shown an interest in OpenRedact. However, attempts to use it often failed due to a lack of the necessary technical expertise or staff resources for operation and maintenance.

Once the funding had come to an end, the project continued to be supported in terms of communications. However, it was no longer actively developed until the BMFTR’s Data Lab, together with the in-house consultancy firm Partnerschaft Deutschland (PD), approached the OpenRedact team and took on an active role in the project.

OpenRedact’s new home: Takeover and further development on openCode

OpenRedact was taken over by the BMFTR’s Datenlabor with support from PD. It was then migrated to openCode and developed further for use by the Ministry. Strictly speaking, the migration is a fork, i.e. a new successor project. However, it was carried out from the start in close coordination and with the consent of the original project team, for whom the handover was an ideal scenario. The original team is now only informally associated with the project. Ownership and maintainer roles lie with the new stakeholders: the Data Lab and the consultancy PD. The original repository on GitHub has been archived; the new main development branch is now hosted on openCode.

In addition to migrating the project, BMFTR and PD made further changes to it. These include, in particular, improvements to the anonymisation function for PDF documents and a redesign of the user interface. Security measures were also implemented, and existing Common Vulnerabilities and Exposures (CVEs) were rectified through measures such as updating dependencies.

As part of these changes, the licence under which the project is published has been partially amended. This was necessary due to the integration of a new library with a stricter licence. Parts of the project are now subject to stricter copyleft conditions: they may not be readily incorporated into proprietary code, and any modification must be shared under the same conditions. In line with the ‘Public Money, Public Code’ principle, publicly funded software thus remains openly available for further use.

What challenges remain? What lessons can be learnt from the example of OpenRedacht?

What is noteworthy about the case of OpenRedact is the collaborative handover process. The handover, carried out in agreement with the project team, and the way the parties interacted with one another should be seen as key factors in its success. For the initial project team, such a handover also means giving up control, which is not an option for all maintainers in the same way.

Moreover, the question arises as to how the project can be secured in the long term under the new owners and decoupled from the responsibility of individual persons and their continued presence within the organisation. In order to achieve this, it is crucial that maintenance and further development are institutionally embedded. It is also important to consider whether an active community can develop around the project or whether an existing community can be ‘brought on board’. A prerequisite for this is that community work, such as the integration of external code contributions, is established.

Finally, lessons learnt from the adoption process should be documented and made available to other public sector organisations — for example, through a public Open Source Programme Office (OSPO). Well-documented processes can make it considerably easier for public sector organisations to get started in similar situations.

While the approach taken by OpenRedact and the Datenlabor does not represent a suitable model for all FOSS projects, it may be relevant for those developing core applications that are signigicant to both the public sector as well as civil society actors and democratic processes. OpenRedact also serves as an example of how sustainable open-source infrastructure can emerge from time-limited funding. However, this depends not only on the funding itself but also on the projects’ suitability for subsequent business and maintenance models.

Judith was responsible for the research of the Prototype Fund between June 2025 and July 2026, while pursuing a PhD on participatory practices in the data governance of data-based projects.