From the sets of stored memory about SPs we may suggest many different inductions that may describe the occurence of SPs, but yet again for the sake of simplicity, it will be better to use the simplest induction possible to describe maximum observable SPs.
This Process of suggesting induction, refutation and resuggesting a more complex induction, is a continued process of creating knowledge.
Stored in Memory...
opp...
Although this statement may be true it can not tell us nothing about the next occasion of breaking glass. If we will test it and we will find that next time we will hear a different sound, we might say that "All breaking glasses so far observed has this specific sound, while the last had another sound"''. By describing the relations between the phenomena in a descriptive mode, we may reach very fast to infinite numbers of description. So yet again for the sake of simplicity, we will try to describe the relations as inductions.
Inductions are relations, assumed to be constant regardless of time or place, between phenomena. In our example, we will say that "Always, when a glass brakes it has specific sound" ("broken glass sound"). Due to the ''Phenomenological cage principal'', we are unable to check that this relation is always true. What we can do, is to try to test it in every occasion the one of the phenomena occur, and to see if it is still stand, or was the induction refuted<ref>Popper, K. (2002). The Logic of Scientific Discovery (Routledge Classics) (p. 544). Routledge.</ref>
When an induction is being refuted, we should try to suggest new relation, this time more complex, to describe the relations that will be true for '''every observation we have seen so far'''. This new induction will be tested again and again, and after every refutation, a new relation will be conjecture and will be tested, when every new induction should describe every observation we had so far.
We should strive to achive the simplest sets of inductions that describes erey observed relations between the phenomena. This set inductions (also called theories) will be called descriptive theories. If the sets of inductions also enable us to manipulate phenomena preciasly<ref>Habermas, J. (1985). The Theory of Communicative Action, Volume 1: Reason and the Rationalization of Society (p. 465). Beacon Press.</ref> it will be called forcasting theories. The inductions are like equations that describe phenomena in more economical way<ref>Ernst, M. (1986). On the Economical Nature of Physical Inquiry. Popular Scientific Lectures (pp. 186–214). Open Court.</ref>.
{{nc|Please better define how we '''mesure simplicity'''|[[User:WinSysop|Tal Yaron]] 08:45, 14 December 2012 (IST)}}
==The Structure of Theories==