# Once again on separation in a parameter

### Y. B. Karasik, Thoughts Guiding Systems Corp., Ottawa, Canada. e-mail:karasik@sympatico.ca

In the 1970s I discovered many new separation principles (see e.g. [1]), one of which was separation between different entities, e.g. different values of a parameter, or different conditions, etc. Later I called them simply separation in a parameter and separation in conditions respectively.

As an example I pointed out to separation between different speeds of calculation. In the 1960s all computers had one processor, which did both calculations and input/output. The last operation was slow and did not allow computers to work fast. Thus, contradiction arose: processor had to input/output data because nothing else would have done it for it, and had to not do it because it slowed down computer. The contradiction was resolved by separation between fast and slow processors. Computers started getting equipped with one fast processor for calculations and several slow processors for input/output.

Although fast and slow processors occupied different places in a computer, the above solution was not about separation in space. The contradictory requirements (has to perform input/output and has to not to) were separated between processors of different speed (i.e. in a parameter). Separation in space was a by-product of this separation simply because two processors could not be co-located.

In most cases of separation in a parameter it is, however, the other way around: separation in space or in time comes first and then separation in a parameter (or in conditions) is picked to just implement the first separation.

Consider, for example, the following contradiction: an object has to be magnetized and has to not be. Circumstantially it is usually about different points in time or in space. Thus, the contradiction is usually resolved as follows: before some point in time the object is made magnetized and after passing this point it loses its magnetic properties. To implement this solution one may resort to Currie point: keep the object's temperature below it when it has to be magnetized and above it afterwards. The later is separation in a parameter, namely in temperature. However this separation is secondary and is intended to implement the separation in time (or in space).

Most instances of separation in a parameter are secondary separations intended to implement the primary separations in time or in space. They are even not usually thought of as separations but as physical effects because most such separations are inherent to them. Examples when separation in a parameter is primary and not embedded into a physical effect are hard to find.

A while ago Olivier Gratzer tried to offer some examples and asked me if they are indeed distinct from separation in time or in space [2]. Back then I did not have time to analyze his examples. Recently I returned to them and discovered that they neither contain seperation in time nor in space nor in a parameter.

Indeed, Mr. Gratzer considered two interrelated contradictions:

1. One has to send over a link more data (in order to deliver a high quality picture) and has to send less data in order to accomodate several channels on a link;
2. One has to maintain several channels on a link (in order to serve more customers) and has to maintain just one channel in order to be able to transmit a high quality picture to a customer;
The solution, which Mr. Gratzder described was to use not one but several links utilizing different parts of spectrum. This is not separation in frequency/spectrum at all. This is separation between such opposites as EACH and ALL. Indeed, the contradictory requirements in the second contradiction are these: maintain one channel per link or maintain several channels. They are separated as follows: we have one channel per link on EACH link but several channels on ALL links altogether.

As for the first contradiction it does not get resolved at all. Just the second contradiction gets resolved. Alternatively, one can view the first contradiction as also resolved but by separation between such opposites as "ACCOMPLISHED" and "NOT ACCOMPLISHED". Indeed, of its two contradictory requirements (send more data or send less data) the first was accomplished and the second was not.

Corollary: usually separation in a parameter comes as a by-product of search for implementation of separation in space or in time. In these cases the problem solver does not even look for separation in a parameter. He or she looks for a physical effect to implement separation in time or in space. Only after the effect is found and applied it turns out that as a result contradictory requirements get also separated in a parameter.

We consider conditions under which a parametr can be used for separating contradictory requirements before they get separated in space/time in a separate artcile.

### R E F E R E N C E S:

1. Y. B. Karasik, Analysis of mechanisms of contradictions overcoming arising in the course of solving scientific and engineering problems, Samizdat, Baku, 1978 (in Russian).
2. Y. B. Karasik, On separation in a parameter, Anti TRIZ-journal, Vol. 1, No. 2, 2002.