david on Nostr: I have always found the term “knowledge graph” to be frustrating bc of its vague ...
I have always found the term “knowledge graph” to be frustrating bc of its vague definition.
Opening salvo in the above article: “Knowledge Graphs are founded on the principles of describing data as knowledge.”
describing data as knowledge? As opposed to what? Data that’s not knowledge? Wtf is that supposed to mean??
And yet … I do not think the term “knowledge graph” is entirely without merit. Its meaning has evolved and continues to evolve over the years, as technology and the tools available to us evolve. At some point we’ll settle upon some more precise and useful definition of Knowledge Graph.
In the meantime, to me, Knowledge Graphs are a subset of graph databases. They are graph databases that have some structure added to them, some set of rules that make them useful.
What are those rules? What is that structure? I think we’re all still figuring that out.
Nostr needs graph databases. More specifically, nostr needs graph databases that are organized into Knowledge Graphs. Nodes for NostrUsers, NostrEvents, NostrRelays, just for starters, with a myriad of relationships to connect those nodes together.
Where will these graphs live and who will manage them? Centralized servers like primal? Personalized relays controlled by plebs? Yes and yes, I think.
How will a Nostr Knowledge Graph be organized? What structure, what rules will we apply to the graph? A certain amount of trial and error will be needed as we figure out how to maximize their utility. Ultimately we’re going to need FOSS tools to maintain structure in real time as new events are added, and sometimes maybe pruned, from your Nostr Knowledge Graph.
What problems will Knowledge Graphs solve for nostr? They’ll help you manage your Webs of Trust in real time. Why does that matter? Because we need to know who’s the most trustworthy to curate our content, facts and information.
Interested in working on graph databases, knowledge graphs, neo4j, personalized WoT relays? Let me know, I’d love to brainstorm some solutions.
LFG 🔥 🔥
Opening salvo in the above article: “Knowledge Graphs are founded on the principles of describing data as knowledge.”
describing data as knowledge? As opposed to what? Data that’s not knowledge? Wtf is that supposed to mean??
And yet … I do not think the term “knowledge graph” is entirely without merit. Its meaning has evolved and continues to evolve over the years, as technology and the tools available to us evolve. At some point we’ll settle upon some more precise and useful definition of Knowledge Graph.
In the meantime, to me, Knowledge Graphs are a subset of graph databases. They are graph databases that have some structure added to them, some set of rules that make them useful.
What are those rules? What is that structure? I think we’re all still figuring that out.
Nostr needs graph databases. More specifically, nostr needs graph databases that are organized into Knowledge Graphs. Nodes for NostrUsers, NostrEvents, NostrRelays, just for starters, with a myriad of relationships to connect those nodes together.
Where will these graphs live and who will manage them? Centralized servers like primal? Personalized relays controlled by plebs? Yes and yes, I think.
How will a Nostr Knowledge Graph be organized? What structure, what rules will we apply to the graph? A certain amount of trial and error will be needed as we figure out how to maximize their utility. Ultimately we’re going to need FOSS tools to maintain structure in real time as new events are added, and sometimes maybe pruned, from your Nostr Knowledge Graph.
What problems will Knowledge Graphs solve for nostr? They’ll help you manage your Webs of Trust in real time. Why does that matter? Because we need to know who’s the most trustworthy to curate our content, facts and information.
Interested in working on graph databases, knowledge graphs, neo4j, personalized WoT relays? Let me know, I’d love to brainstorm some solutions.
LFG 🔥 🔥