Wikipedia does not cover the entirety of human knowledge - it's limited to long-form, text-based representations of encyclopedic topics on subjects well-covered by reliable sources. There have been recurring discussions about covering new types of knowledge (oral history, collaborative video, genealogy, fact checking, 3D models etc. etc.) but they never went anywhere, because creating new projects is a lot of work and the outcome is unpredictable (some things only work in practice), so the Wikimedia Foundation was unwilling to take the risk. While a new project proposal mechanism exists in theory, it is largely defunct - the last new Wikimedia sister project was Wikidata in 2013, the one before that was Wikiversity in 2006.
If we want to make meaningful progress on our vision of curating the sum of *all* knowledge, and truly become the infrastructure of free knowledge, we need to overcome this barrier. As Wikipedia's example has shown, it's hard to predict which projects will be successful - to find the ones, we need a project proposal and incubation mechanism which can fail cheaply. We need infrastructure for low-effort, low-risk experimentation with new projects - a "wiki nursery". A system where new wikis can be created at minimal cost, technical and social experiments can be performed on them without undue risk to other projects, and they can be discarded if they prove unsuccessful. Other organizations have used such mechanisms with great success (Wikia has tens of thousands of wikis; Stack Exchange has nearly two hundred sites).
After making the case for such a system, the presentation will discuss potential technical and social problems, and ways around them.
The presentation is in the context of a planned proposal for a software project (the sister project incubator). Outcomes will be raised attention for the problem of project type innocation and its connection with knowledge gaps and knowledge equity, and feedback about the feasibility of the proposal and potential pitfalls.