Much of the formal work leading to BNT was the result of Swindler (1950) who laid the foundation of the general theory by developing a concise definition of Browness:

"For any assumed real number, N, utilized within an equation, there exists an associated value, Browness, between 0 and 1 which is a function of the uncertainties in the boundaries and the constraints of the problem to be solved as related to the real system (reality)"

Swindler went further describing Brown Numbers as a type of fourth dimension which others have described as the "Brown Space". Extensions of this aspect of Brown numbers have been used in quantum physics and surreal arguments about the “color of the sky in your world”.

Conman (1961) proposed a definition of a Brown Number as one with a Browness (Br) of 0.2 or greater.  This definition is empirical, based on the observations of Conman and Pick (1959).  Conman termed a Browness of 1.0 "saturated Browness" or Bs. Conman did not believe there were and truly Bs values and that all values had at least some Browness.

Fudge (1965) refined Swindler's work and published a collection of equations of Browness (also referred to as Brown Functions) concerning certain financial transactions.  This monumental work in the realm of Brown Numbers led to the development of many other equations of Browness.  Fudge reportedly published his equations in a standard form with a table of constants for various applications, although only fragments of the original paper remain.  These constants are referred to a Fudge's Constants (or Factors).

A major contemporary use of BNT is the extrapolation of a limited set of data.  Conman (1961) coined the term extensionism and was the first to apply BNT to data trends.  Conman reasoned that the higher the risk or uncertainty the corresponding higher possible gains.  Because Browness generally increases as a function of risk and uncertainty, Conman concluded that the larger the Browness, the larger the possible gains.  From this logic, Conman proposed the following Laws of Browness:

I) Absolute facts have zero Browness and random assumptions have a Browness of 1.0.

II) Values with higher Browness are more desirable for greater gains, unless the value conflicts with an absolute fact (reality).

III) Any subset of a collection of data may have one or more trends which may be extrapolated without regard to the remaining data to achieve maximum Browness without conflicting with an absolute fact.

(c) 1982, 2009 Jorge Branche, Jr.