Juha Korpela gets it, everyone is arguing about the place to store context/semantics or what format, but hardly anyone is talking about what I think is the most interesting part - HOW we gather it.

• How we support tools for humans (which I’ve written about here: dataprodmgmt.substack.c…)

• How we extract semantics from existing corpus using AI (methods exist, but are they user friendly enough to get traction?)

• How we connect external semantics to internal semantics in sustainable maintainable ways

More content on this please! Less content debating who gets to own the meaning (ironic) of the buzzy term context graph.