Hello everyone,
Thank you for your interest. We are working hard with the Free IT Foundation to release a stabilized beta version. If you leave us your email address, we will notify you as soon as the code is published under GNU AGPL v3.
We are facing a triple challenge.
1. New features from the 7 Billion Urbanists network
The Open Urbanism Foundation wants to introduce new features based on exchanges with the 7 Billion Urbanists network, mainly improved ergonomics and a dual tagging system (project tags and community tags).
2. Multigination, European research project (2024–2026)
Multigination is a broader European research project bringing together several research partners and the cities of Winterthur (Switzerland) and Başakşehir (Turkey). Within this project, the Open Urbanism Foundation's specific role is building a dashboard that lets cities, research consortiums, and organizations deploy the tool autonomously on their own projects. Each project owner can either draw on the shared community database (categories, tags, items, questionnaires built up from previous projects), build everything from scratch, or, as is most common, create a hybrid of both. Any new elements introduced by an organization can then be fed back into the community database, contributing to a broader dynamic of international cooperation.
3. CoreSpace, European research project (2026–2029)
CoreSpace is another, wider European research project currently underway. Within CoreSpace, the Open Urbanism Foundation's specific role is to prepare international comparative analyses, including retrospective ones, and to integrate artificial intelligence in a deliberate, reasoned way, along two dimensions:
-
Analytical (mainly quantitative): giving project owners and contributors an initial layer of analysis on factual elements, such as timing, locations, number of participants, number of collages, and typologies, across the whole dataset or by location.
-
Generative (semantic): exploring themes, objectives, and spatial arrangements proposed in collages, comments, and questionnaires, to help both professionals and civil society push their thinking further, grounded in the factual data produced through the process.
Multigination and CoreSpace are two distinct European research projects, each with its own scope, timeline, and partners, and each broader than the Foundation's specific contribution described above. They are nonetheless designed to be complementary: Multigination lays the operational groundwork (dashboard, autonomous deployment, community database) that CoreSpace will build on to add comparative and AI driven analysis. Progress made in one project directly strengthens the tools and data structures used in the other.
Our approach to "reasoned AI" for open urbanism
Conceptually, we see AI as a support tool for professionals and civil society when analyzing results. Navigating and drawing insights from hundreds of responses that combine images, text, answers, and tags is extremely time consuming. Linking the structuring of this data to the raw data itself creates new opportunities for professionals and citizens to genuinely listen to one another. The collage tool empowers creativity by reducing the gap between those who can draw and those who cannot, and it remains strictly a human tool. The structuring tool is accelerated by AI, but it stays permanently verifiable through direct access to the raw data.
Technically, thanks to our partners Infomaniak and Free-It Foundation, we use servers based in Switzerland and open source, open weight models that are sufficient for the kind of analysis we need, without joining the race toward ever larger frontier models that require increasingly massive data centers. On the contrary, the models we use should allow for self hosting in the short to medium term.