The basic starting point (as discussed in concept graph readme? or DCoSL readme? or glossary? or create a DIP-0* ?) is to represent data as a graph, with nodes and edges. (State elsewhere: Graph databases like neo4j could be considered but are not necessary and not necessarily recommended and probably not beneficial until later stages with large concept graphs.)
The special role of graphs in the DCoSL protocol is discussed in concept graph.
DIPs (DCoSL implementation proposals) are categorized into three blocks: the core protocol, the concept graph, and the grapevine.
a few fundamental ideas that require explicit consideration
how data is represented and organized
how data is curated
publication and retrieval of data to/from the nostr network
a state of enlightenment where the implementation of DCoSL is itself crowdsourced using DCoSL.