Science is recursive. One hypothesis (approximation) superseded and falsified by the next. It leads to better and better answers.
It converges. It becomes more complete.
The model doesn't have to be precise replica of reality. It just has to be "good enough" to make predictions.
The point is that the word "coffee" was interpreted by you and you understood it. It triggered action.
You converted my declarative knowledge (make coffee) into a procedure (boil kettle, put coffee etc.).
Or the alternative procedure utter the phrase "you are off your tree".
Self-driving cars. Machine vision. Automation.A_Seagull wrote: ↑Sat Dec 01, 2018 1:55 pm And computers are only perceived to follow instructions, it is only the perception by a person that 'grounds' its response in 'reality'.
It is only people that have a model of reality (at least so far in technology). Computers and modern robots are no more than machines in the same way that a plough is a machine.
Robots are beginning to do more and more tasks which were previously considered in the domain of "humans only".
The distinction is no longer in what humans can but robots CAN't do. The distinction is "is there such a thing that robots will never be able to do" ?
Everything technology is able to do right now is defined IN language. That means we are TRANSLATING procedural into declarative knowledge.
http://unt.unice.fr/uoh/learn_teach_FL/ ... ategorie=3