—
Point control is a set of incoming quantities as commands received from the outside through control and measuring devices (sensory organs). Each command is a definition (and the definition is a command), so geometrically each command is a straight line segment, where one of the ends of the segment belongs to the subject and is its own parameter. The second end belongs to the object and reflects the amount of difference from the subject. Since the subject cannot form a straight line segment with himself, the external points must be set in advance. The assignment of such points is called an organization.
The cases under consideration contain one irritant, but what if there are several of them? So we need to move on to solving the problem of choosing a point from the set for interaction, which will become a matter of sorting. Although common sense suggests that there is the simplest solution, it is enough to add up all the available amounts of excitations and divide by the number of pathogens. Take and divide where the average temperature in the hospital will become the desired amount of excitement. But let's not look for easy ways. The task of choice is to determine with whom and how one can interact (and with whom one cannot). It can be tough, it can be an algorithm, the options can be very different. Usually such a predestination is called an organization. The latter is the prescription, established connections for interactions, established rules and algorithms, and so on. They may be different, for different tasks, but there will still be something in common, some universal techniques and methods, or so. To solve the problem, we outline the introductory ones. Imagine a point around which there are several other points, from any of which you can form a straight line and a straight line segment (definition). But a point can only interact with one point out of several. Accordingly, the rest need to be eliminated somehow. In order to filter out something, you need to set differences, some criteria by which those who do not meet the criteria will be excluded. And there are several ways to do this.
—
The first and simplest principle is the principle of the largest or smallest value. Imagine a set of points, each of which has a certain value in itself. Whatever the values, but among the many there is a minimum and maximum. If the interaction should be with only one point from the set, then the points with the minimum or maximum value will filter out the remaining points with a different value.
For example, choosing the point closest to itself as the choice of the smallest value. There is some distance between the defining point and the rest of the set. Which is a set of quantities. In this case, you need to choose the minimum value from this set. Which will turn out to be the smallest distance as the smallest length of a straight line segment. The interaction is formed by irritation and arousal caused by it. This option does not guarantee the greatest amount of excitement from this point. But, firstly, sometimes it is not necessary. It is not at all necessary to focus on the point that causes the greatest excitement, it is quite possible to sort only by the distance to the points, if we think in advance of the interaction forced, without irritation and reciprocal excitement. Or we do not need exactly the greatest excitement, it is enough just to have at least some with the nearest point, because it is the closest. Secondly, if it is still necessary, then it is not necessary that the points have a valid definition in the coordinate space. If the surrounding points have an abstract definition, which is a constant in magnitude, then they will differ from each other only by the distance between themselves and the defining point. Accordingly, the closest point will provide the greatest excitement, the rest will be eliminated.
A variant of the solution is the choice of the most pronounced of the surrounding points in magnitude. What kind of property it has the most pronounced is not important. That's why it is a primitive, that the surrounding points have the same property that the defining point also has (otherwise they will not be able to form a straight line through each other and a segment on it). Here it is similar to the distances in the coordinate space, only now it is the choice of the largest length of a straight line segment (points in the value space also have distances between themselves as straight segments with some length). There is an option to sort only by the largest value difference, with whom it is higher, and with that there will be interaction. The amount of excitation is also secondary here, it does not necessarily have to be the largest of the available ones. There is a variant of sorting by the largest amount of excitation, but with abstract distances that are constant for all points. Therefore, the points differ from the defining one (and from each other) only by the distance in the space of values. And although sorting is performed as a choice of the largest amount of excitation, but in fact it is the choice of the largest value difference. Since the difference in values hides the difference in the severity of some property, then this type of sorting is convenient when it is necessary to calculate exactly the maximum severity, ignoring the distance to it. For example, if it is dangerous for the point in some way. Accordingly, the interaction will be only with this point, all the others will be eliminated, even if the amount of excitation with them is higher.
The two listed options do not negate the opposite, when the largest distance is being searched in the coordinate space, and the smallest in the value space. This is a case of forced search, although it is difficult to say why this is necessary in practice. Because in practice, the interaction is formed from irritation and response arousal, which has a certain magnitude. In this variant, the point that causes the greatest excitement in magnitude is searched for. The surrounding points are similar to each other, and they do not need to differ from each other in any other way. Because they will differ from each other by the distance between themselves and the defining point in the spaces of coordinates and values. They can be at different distances or have different differences in values, so each will cause a different excitation in magnitude. Accordingly, the solution to the problem is sorting the surrounding points and choosing the one with the maximum amount of excitation. Either because it is the closest, or because it has the greatest difference in values, or all at once. It is this option that can be considered standard. This is a search for the largest amount of excitation with valid definitions in the coordinate and value spaces. This means that a computational operation will be performed for each point, and only one with the maximum value will be selected based on the results. Even if some points will have the same distance to the defining point, they may have a different difference in values. Or vice versa, the same difference in values, but different distances. Only in special cases will everything coincide, and hence the amount of excitement, which will lead to a collision of a donkey and two piles of hay. It is unclear which to choose from the two. In other cases, this principle does not need anything, because everything is already there.
The above reveals the essence of both sorting and organization — these are customizable links. To make a choice, you need to set an algorithm that will guide the choice of what exactly you need to choose from the set. In this case, no one prescribes to the point any specific choice with which point it is necessary to interact, and in no other way. On the contrary, a point is prescribed to interact with any other point, but under some condition. In this case, provided that the value of the other point is the largest or smallest of the available ones. All together, there is an algorithm as a condition or a set of conditions. In the simplest case, the condition requires itself to be true or false. Therefore, search and sorting is a true search, a search for compliance with a given condition. In a more progressive case, the condition may contain values. These conditions, whatever they are, are internal. This is part of the algorithm, but the algorithm also has external logical conditions. Where the condition is a trigger to start the search. For example, if A = B (more, less), then you need to start sorting to find the largest or smallest value. The external condition may or may not be. In the latter case, sorting is performed by default, it does not require a specially specified start condition. By themselves, such conditions are not a selection criterion. The selection criterion is an internal condition, the interaction is considered true with a specific point, if, for example, its value is the largest or smallest. In other words, there are conditions that define interaction as such, and there are conditions that define the choice with whom to interact.
Thus, you need to specify the differences through which one point will be cut off from another. In the first case, the points differ in distances and values. Thus, they are already individual, they can already be distinguished from one another, they are already cut off from each other, and so on. The strengths of this method of organization are the ability to prohibit interaction with some points through clipping. For a variety of reasons, and the principles here are a combination of such reasons. And vice versa, it is allowed to interact with some points through sorting. Moreover, through algorithms, it is prescribed to interact only with some points, and not to interact with others. The weak points here are a continuation of the strong ones. It may happen that two or more points will be indistinguishable, the same when selected. And the interaction can only be with one. It may happen that some undesirable interactions are not cut off, or desirable interactions are cut off. And also for various reasons. And so on, that it is necessary to anticipate in advance and be careful.
The first principle is the simplest. It rests entirely on definitions. Either the greatest excitation, or the smallest distance, or the greatest difference in values. You can still come up with something on the same principle, when the choice is made in favor of a larger or smaller value. The value itself comes from anywhere. It can be sorted among the available ones, or you can do it a little differently.
The second principle is the principle of threshold values. The latter can be either on the one hand, when the set of points is cut off only along the lower or upper boundary, or on both sides, when the points and their values are required to fall into the range. Again, imagine a set of points, each of which has a certain value in itself. Whatever these values are, there will be points among the set whose value is less than the minimum set or greater than the maximum set. Such points will be eliminated.
This brings us back to the solution of the problem where you need to choose some point from the set. It is clear that such a principle is no longer suitable for choosing interaction with only one point, because many points can fall into the range, but only one is needed. But sorting and selection is used not only to organize interaction. Sometimes you just need to thin out a lot of points to some modest amount. And then perform another sort to select only one point. The principle of threshold values fits perfectly here. He does not so much tell the point with whom and how it is possible to interact (with anyone who falls into the framework), as he talks with whom and why it is impossible to interact (anyone who does not fall into the framework). Because they could not cross a certain threshold in their values. Or their values exceeded a certain set ceiling of values and also in the span.
If, however, there is a need to select only one point from the set for interaction, then the range of values should be very narrow. It is desirable that the upper and lower boundaries close to the same value. Then the sorting will be reduced to finding a point whose value corresponds to the specified one. This is an example of tightly organized interaction with strictly defined points. If the point changes its value, the interaction with it will stop, because it will fall out of a narrow range. This technique is typical for a rigid network organization, where interaction occurs only between such points that correspond to a given value. Any other points with different values are cut off here. As a consequence, in such an organization, all points are likened both to each other and to a given value.
In a more advanced version, a given value can float in a certain range with some frequency. If the points float after it, then the set of points change their values to a given one. As a result, the ersatz of rigid organization is preserved, the magnitude of the points is constantly changing relative to the external observer, but does not change relative to the internal one. Because the points at once changed their values to new ones, which did not add any differences between them. As they were likened to each other, so they remained, only now according to a new set value. Periodicity implies a wave as an oscillation, in this case the whole set of points will oscillate after a given value. If the points do not change their values, then changing the set value will periodically turn off some points from interaction, but connect others. Today you need to choose a point with one value, tomorrow the fashion has changed, you need to choose a point with another value. We are well aware of this, because society is organized approximately like this. As a result, the set of points will be constantly updated with a new value. Some points will fall out (completely or until the next wave), which is also a sorting option.
The wider the range, the more points can be involved in the interaction. But we remember that interaction can only occur between two points, so we still need to thin out further. For example, there may be several threshold values, when points can be thinned first by one criterion, then by another, and so on, until there is only one point that meets all the necessary criteria. How to pass through a sieve.
—
Previous chapter — Conducting excitation in the value space
—
Next chapter — Psychological organization of points
—
Comments