@joonhyeok_ahn @nickanderson So far software has helped us
1. store our knowledge non-linearly associatively
2. keep a consistent system (templating, ontologies, typed data structrures)
3. is helping us with discoverability (keywords, n-categorization, heterarchies/poly-hierarchies, semantic search and tagging)
4. the next thing is help us keep consistent workflows and consistent workflow steps?

certainly #aHypothesis of mine.

#AHypothesis - open-source pkm tools need more onboarding. The fact that some initial strucure and opinion is found in the more "commercial" tools actually helps people get comfortable with them.

#AHypothesis - we need #discoverability and we need #hypotheticalStructure . We talk about letting structure emerge but don‘t provide tooling (besides „if you can‘t spend ten minutes processing it don‘t do it) to build structure later. Software tools that help end up being more comfortable.

#AHypothesis for #pkm as A Thing in software - No single software lasts forever. So we must
1.) ensure there‘s a common understanding of *information relation, presentation and manipulation* concepts
2.) that there is a healthy #FOSS ecosystem that captures all of these parts in a modular manner, so that overall, all the domains of material remain captured

The diversity of developers and modularity of individual implementations ensures that overall systems remain stable for decades.

#AHypothesis of #pkmDesign - the power of modern PKMs is captured the most in three basic file formats:

- free-form flowing text, usable for outlining with bullets and long-form writing. Ideally it supports increased degrees of advanced formating. Individual segments can be elevated into structured data.

- OPML-like structure-first systems for personal databases. Stores things with high native categorization.

- Free-form Canvas tool for connecting other two file types and free drawings.