How do you conclude preferring apples to oranges involves limited information? Please explain.
1. I thought you might not like the term I first used (weakness), and that calling it a "limitation" might be more acceptable - closer to a moral neutral.
2. I'm assuming the being makes a choice of fruit because it is beneficial - for whatever reason: pleasure, nutrition, to employ fruit growers, whatever. IOW, it's not an arbitrary choice, because I don't see how the terms "preference" and "arbitrary" logically fit together.
3. I'm assuming it's possible for this amazingly gifted being to maximize the benefit of its choice. It is within the ability of this being to gather all the data needed and to process that data to determine a maximum. Further, the necessary data exists. Anything else would constitute a limitation. Or, if there are 2 maxima of exactly equal value, we're back to this being an arbitrary decision, and therefore not a preference.
4. If there are 2 different beings, the maximal choice of eating an apple or an orange might be different for them. This might seem to create a conundrum for answering the question if both were perfect beings, so I approach it this way: separate the mind from the body. A perfect mind will make the choice for a body, regardless of which being the mind and the body belongs to. Since there is only one maximal choice for that one body, any mind that happens to be perfect will find that maximal choice.
Therefore, all perfect minds will make the same choice for a particular body and cannot be distinguished based on this choice. The only thing that would distinguish them would be one of the limitations listed above.
Along this line of reasoning, I did a little digging around suspecting that someone might have already pondered this issue of preferences. Sure enough, it's something that comes up in the field of computing. Computer science is interested in the idea of the "heuristic", which is defined as an algorithm that will "find a solution which is not guaranteed to be optimal, but good enough for a given set of goals."
Heuristic - Wikipedia, the free encyclopedia
The reason for this interest is that computing a decision based on
all the information can be very slow, so a heuristic seeks a decision based on
limited information because it can come very close to the goal in a much shorter time and get a "good enough" answer.
It just so happens that computing a decision based on limited information is called a "preference heuristic". The idea is to bias the algorithm in some way so it will make a decision even though the information is incomplete. You can find papers on the topic such as this one:
Using preference based heuristics to control abductive reasoning - Springer
Or, there is a preview at Google Books that lists some of the types of preferences (biases) these algorithms use:
https://books.google.com/books?id=s...w#v=onepage&q=preference is heuristic&f=false