That sounds nifty. This lossylossnessness probably is a good way to think. What I would say, is you wouldnt want to lossless first, youd have to be lossy first, because once its lossy u lose all your literals, but I imagine you already know that, and u would get more max compression out of it, but I only care about lossless, context loss is not good when its logic, its only good if its media and senses dont notice it.
I have another idea about compression - not involving logic, but media, and its like a neural network which has so much animation store in it, (and it can be quite large! thats ok, because im only reducing the literals, so the zipping program can be quite large and i still havent defeated the purpose) and then the videos all convert to it - and basicly what im doing is im reducing permutations by making everything more the same. The idea isnt complete, but the main idea about it, is 32 bits has 4 billion potential movies in it. (almost more than have ever been made.) but its non mappable, just with 32 bits. But the idea is you want to reduce the amount of context, then there is less pipeways to worry about. But the idea doesnt work yet. ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-Me26fc7f89225acd2f6d21292 Delivery options: https://agi.topicbox.com/groups/agi/subscription
