@Matt, if his code worked and **then* *we read his book, that wouldn't make us understand his book any better. It may make you **sure* *his explanation is true, but if the explanation doesn't prove itself without requiring code, then that means the explanation doesn't hold up itself and doesn't explain all things to make you say "ah". Code takes a long time to make and mostly does only what you *already knew* it'd do.
Matt, you work with predictions, an agent can be very sure if he has lots of context backing his prediction. I am sure of my AGI design. I don't need code. I can fully explain it. What I am missing, I have _no clue_ how it works. And to code something the idea/egg comes before the hatch. I can literally get out my paint program and make discoveries, using my visual knowledge while draw a network. My brain is able to make a lot of really good predictions; I have a good model of physics/ our world, or at least the A.intelligence domain. ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0e5ebbcbbc24ea9f-M7786e52b8cb1f33d74419357 Delivery options: https://agi.topicbox.com/groups/agi/subscription
