graphinius

graphinius copied to clipboard

there are different data structures for different use cases, which is legit, BUT

• we need something like a standardized compute pipeline
• starting with ID(s) or single Nodes
  • --> expanding up to larger sets
  • --> extending to a set of sets
  • --> expanding to a set of larger sets
• which in each phase only accepts a certain type (data structure)

• and leads / expands to a well defined type the next step up