| Some thoughts. I should give this topic some time to brew an answer, but at some point we ought to ask ourselves, why don't we have such a thing? If we were to construct this database, should it not be on the level of the Human Genome Project, but open sourced? Should it not be more fundamental to have this easily transmitted database, maybe even Unified Design, than Defense Budgets and the Billions that must've gone into the Antarctic GitHub vault? Ideally, the platform of world-wide human ownership of this project should have been initialized and set up for us to complete by, like, I dunno the U.N. or like funded by all Governments and Corporations as a good faith initiative? Tangentially, let's also suppose that we are not alone in the Cosmos is probable enough to merit effort to enter the next phase of simply sending a Clarke-like DNA/Planet position plaque? Should we not prove our evolution and broadcast this supposed database into space in a loop? The Ascension Beacon? The excellent comment @speedghost got me thinking about the human interface with AI from a different perspective. Why is the AI public focus on language models, but not knowledge models, how to think about thinking. Is that not "the trick"? Could Humanity not use this itself to better learn to learn? How to categorize, or order, knowledge about knowledge (metadata) is "thetadata" (thinking data)? Thoughts about a potential Human Knowledge Database. Wikipedia is a mess and is not fit for this purpose in many ways, but the main one is contains sparse info on how to do anything. If the goal is to make the database the most efficient (small size) as it can be, and Unified Design, for all Humanity, we might need some kind of new "motion illustration" format that consists of pictures that are flat files but can be post-sequenced, to create a "movie" with the lowest possible file size, and using vector graphics only? Brainstorming. That makes me think the database should be some sort of flat file database that is also indexed, and that index can be easily Human parsed or DB software parsed. Quite a feat for 100GB. That is not my lane, maybe it already exists. I'm holding out hoping there is literally one person out there who already has the seed for this, maybe even already has a project going. |