We combined both sessions because of common interests.
The topology of current internet platforms and infrastructure is constraining our capacity to grow as individuals and to advance as a collective. We are in a need for tools that can interoperate and allow us to organize, share, and fork knowledge transcending these limitations.
Prior to and through the camp, several of the participants had met already. There are quite a few projects tackling these issues from different perspectives.
- Guillem Córdoba (@guillemcordoba) _Prtcl
- Vincent den Boer (@ShishKebab) WorldBrain and Storex
- Akshay Kanthi (@aksanoble)
- Roman Khafizianov (@requilence)
- Justin Maire (@JustMaier)
- Jose Ospina (@pepoospina) _Prtcl and CollectiveOne
- Marc-Antoine Parent (@maparent) hyperknowledge
- Anton Pronkin (@pronkin) Anytype
- Oliver Sauter (@blackforestboi) WorldBrain
- Xavier Vives (@xavivives) Interplanetary mind maps
We were a large number of participants with a wide and abstract topic to cover. The timeframe of the event it barely gives us time to expose our personal perspective with a high level of abstraction.
We agreed that defining a data model is the hardest and more critical element to resolve, but to move forward we need to outline clear use-cases. We decided that a first step in this session was to brainstorm use cases that an interoperable data model for knowledge management data need to cover in order to work. The notes should provide ground for further discussion on how exactly a model could look like
So far we collected the following input:
- Traceability > causality > sequence of actions (e.g. merge history)
- Flexibility in interpretation, aka. "views"
- Connect anything with anything. (e.g. with a graph)
- Ability to break or modify hierarchical social structures > fluidity
- Hierarchy of context > local hierarchies
- Federation > choral explanations
- Deduplication, (local) merging and forking of arguments, ontologies, facts, reasonings and concepts.
- Support for public and private sharing and encryption
- Extensible metadata, even as invisible payload
- Indexability & Searchability > ideally with different indexes
- Interoperability with existing standards
We have created a Telegram group to move the conversation forward.