Sunday, October 31, 2010

(?)?

The idea of giving such a title to this post isn’t entirely mine. It is inspired by someone else’s idea of naming something else( :D :D ). What it means is something the reader has to imagine after reading the post :P . This post is not related to the original PvsNP problem; but it is certainly inspired by that problem. It is about the question of questions.

A question, has a shape. It has two components.

1.) A solution space. (i.e., one should know how the answer ‘looks like‘)

2.) A verifiable condition.

The question is to look for some element in the ‘solution space’, which satisfies the condition. As an example, the question 2-3 = ? can given a shape.
1.) the solution space is the set of integers
2.) the condition is, 2+x=3.
That’s rather numerical. But the scope of this shape of a question is much wider than questions related to numbers. The space and the condition are much more abstract in many useful cases. An essay; everyone knows how to check for the condition. But what is the solution space? You could use “the set of all essays”, if the meaning of essay is known. Or, the “set of combinations of the 26 letters, the space( ) and the other symbols used” :D . That looks awkward. However, the point is that one should know what the answer looks like or what are we looking for. That is the job of the solution space. The two examples should be read and forgotten, the crux of the story is yet to come.

So, why not always take the so called universal set as the solution space, and reduce the structure of a question to just a condition? (which is what most of us think of a question as). Well, the universal set, if it exists (no, it doesn’t!) doesn’t tell us anything about how the answer looks like. A solution space can be any big. But it must tell us what we are looking for. By the way, for those who were surprised at my earlier remark, the universal set does not exist. One can not create something out of nothing. Assuming that there is something which contains everything results in a paradox, called the Russell’s paradox. All it means is, ‘you cannot put all thinkable objects in a single set’.

Constructing the solution space turns out to be the major issue in building a question. Most questions which seem to be unanswerable are so simply because they don’t have a solution space(I mean, we don’t really know what we are looking for!). Just an attempt to construct a solution space resolves many of such queries. So, whenever a perplexing query comes to mind, one has to stop and think what am I looking for

As it turns out, it is a very non-trivial job to build such a structure to the queries of the human mind. As a matter of fact, the problem of finding such structures is itself a structured question!. However, in this case, the verifiable condition is given by the satisfaction of the mind. That makes it somewhat different from ordinary questions. In fact, it makes it interesting(=less boring :D ). Figuring out what our mind is looking for forms the core of thinking.

What does one do after structuring the query? nothing! :D . “The real job of a mathematician is to get equations, not to solve them!”. Solving them is the job of a computer. whatever needs to be done next is too ordered to interest the human mind. However, it seems ‘finding’ the answer turns out to be either too trivial or unimportant. So, before asking “how can a man pass through a wall?” one has to stop and think what exactly is our mind looking for, and in many cases, such an attempt alone can resolve the query.

wordpress link