Modular articles allow you to work with the purest form of a "concept" as defined by the author domain/global.
it gives you the coordinates to the raw materials that compose a full text. Now it is up to you to identify the most meaningful bits as defined by you, and sit there for a bit.
What are the coordinates? Text embeddings.
Subjective, transitive trust graphs + modular/modularizable, forkable articles = the raw materials to properly rebuild the scientific publishing world.
There are blockchain solutions too, but I prefer the subjective reputation route for reasons that becomes obvious if you think about it for a moment.
Yeah i keep hearing about blockchain solutions for knowledge graph to mantain and verify when and who made edits. Giving an honest attempt to understand, I haven't been able to understand why blockchain would be needed.
80% iq and whatnat i guess 😂💁
if it doesn't resonate or can't be explained to me as simply as that first nostr doc announcing the functionality - i'm not going to have the motivation to be diving into it until i get that spark.
You are correct, there is no need for a blockhain. (Unless you want consensus + global state; but it is self-evident to me that you do NOT want global consensus on trust and reputation. Last place you'd want that!)
You do want global consensus on who wrote what. It will probably work the same way proof of work timestamps already work, I think, maybe?
