Please improve these comprehensions (was meaning of [ ])

[Ben Bacarisse]

Chris Angelico <rosuav at gmail.com> writes:

> On Thu, Sep 7, 2017 at 5:59 PM, Marko Rauhamaa <marko at pacujo.net> wrote:
>> Dennis Lee Bieber <wlfraed at ix.netcom.com>:
>>
>>> On Wed, 06 Sep 2017 10:37:42 +0300, Marko Rauhamaa <marko at pacujo.net>
>>> declaimed the following:
>>>
>>>>
>>>>Which reminds me of this puzzle I saw a couple of days ago:
>>>>
>>>>   1 + 4 = 5
>>>>   2 + 5 = 12
>>>>   3 + 6 = 21
>>>>   8 + 11 = ?
>>>>
>>>>A mathematician immediately comes up with a "wrong" answer.
>>>
>>>       I'm coming up with "96", on the basis that the "+" is a
>>> placeholder for a non-standard operation
>>
>> That's a mathematician's "wrong" answer. Stefan Ram's answer is the
>> intended one.
>
> Ian's answer has better justification. I'm going to support "banana".

Solving puzzles like this should come with a pleasing "ah!" moment which
usually comes from finding a simple rule or explanation.  An arbitrary
answer is always possible, but it is rarely a rule, and although
arbitrary rules are also possible, they will rarely be simple.  (On
average a rule will need at least as many symbols to be described as the
puzzle itself[1].)

The trouble with this puzzle is that has at least two answers that are
simple rules and, to my mind, neither has a pleasing "ah!"  associated
with it.

[1] Obviously this is not formal but it could be made so by reference to
algorithmic complexity.

-- 
Ben.