In the beginning of the year we were asked to write about the game of “Life” created by Conway. I was a complete skeptic, believing that it is near impossible to create life though a computer program due to the fact that there are so many exceptions and unpredictable actions. I still believe this but after reading about Metacat and playing around with it, I am starting to see a shit in my beliefs. Metacat puts up a good fight in convincing me that a computer could really think and account for life.
Metacat with the combination of copy cat has done a good job in representing a computer program that seems to think. With in the programs themselves there are sub-programs and rules that take in many exceptions, such as pattern clamping and temporal traces. With all of these sub-parts it seems that Metacat is indeed “thinking”. The steps that Metacat takes to come up with an answer can be somewhat parallel to the way a human brain works in thinking, the only exception being the speed at which it processes.
It’s amazing if you think about everything that goes on in our bodies without us even knowing.
If someone say to a joke like “A man walk into a bar, he should have seen it”, we as humans have the ability to hear that and understand the meaning outside of it literal contexts. But for a computer to do this is a little harder. Metacat can make analogies and process thing out of a completely literal state, but only to some degree. It will work on an analogy and come up with answers that are not complete literal answers, but if a problem is too hard it seems to default to a direct literal answer. This was shown in the abc to abd xyz to xyd. It seems that the pattern was to take the right most letter and turn it into the letter that follows it, c went to d. But with xyz example there was a snag, z has no following letter and the rules state that there is no loop in the letters. So the computer had to think and use all it sub parts to come up with another way to view the pattern and problem. Which was the right most letter changes to d.
“When humans make analogies, we not only construct mental mappings according to constraints, we also understand the meaning of the concepts connected by these mappings.”(Marshall, 5) We make analogy to connect to realms of our knowledge to strengthen our learning. But when a computer makes and analogy sometimes that transfer and strengthening of knowledge is not really there. It was cool that Metacat has the ability to know that it had done the problem before. But what I found even more amazing was the fact that it could recognize that it had done similar analogies. I did a cat goes to bat and then a sat went to bat and it recognized that it was similar to the cat to bat problem.
Metacat definitely gave me some good insight. I found myself trying to connect what I was seeing on the screen with what might be happening in my brain at the same moment. I think humans in general take for granted just how amazing our brains are. Some of my classmates comment on how they got bored while running programs. To me I think the bored comes because we are typing in problems that we already see a pattern too. Before typing in the letters I think we all feel like we know what should be the out put. Metacat is seeing the problem for the first time and is going through the whole thought process that we already did. I feel that we just get inpatient.
All I can say is that I for one am very thankful that I do not have to mechanically think about every step in a process when given a problem. I am happy that some how my brain just works.