Kevin Simler

@sarahdoingthing @Meaningness 2. Knowledge best understood as a process of making models better align with reality.

@sarahdoingthing @Meaningness 3. It is a physical process. Computers and other non-humans can (theoretically) generate knowledge too.

@sarahdoingthing @Meaningness 3... E.g. natural selection is a knowledge-generating process, encoding its knowledge in genomes.

@sarahdoingthing @Meaningness 4. The most important/broadest thing to say about this process is that it works by conjectures & falsification

@sarahdoingthing @Meaningness 4... In other words, the only way to get knowledge is to weed out non-knowledge, i.e., ideas that don't work

@sarahdoingthing @Meaningness 5. Whenever a system is "learning," this kind of trial&error must be occurring (at some level of description)

@sarahdoingthing @Meaningness 5... even if there are other ways to describe what's happening (e.g. phenomenologically).

@KevinSimler @sarahdoingthing @Meaningness Problem is both confirmation & falsification give us *some* info, so Popper view seems awkward.