How to define?
There are certain rules that must be followed when we define a concept. I met these rules a long time ago. Thank you father! Once he gave me a book by Saharny L.V. How our language works . The book is written for children and therefore very clear and simple:
Since, as analysts, we have to give definitions all the time, it was nice to remember these rules.
Why am I so seriously considering the topic of definitions? Because I regularly meet definitions that cannot be used in practice. And I do not see a trend towards improvement. To begin with, we will analyze the term process by analogy with the way we have discussed the term system in the last article.
I have seen many process definitions, each of which contradicts the other. Here is one of them:
A process is a set of consecutive actions aimed at achieving a certain result.
')
The reason for the contradictions is that with one term “process” means a lot of different concepts (I counted up to 30 at one time) and the definitions are trying to give him only one. If they were all listed separately, there would be 30 definitions of different concepts, denoted by one word.
Well, if the word has only a few meanings that are easy to read from the context. For example, the meaning of the term "spit" is easily guessed from the context. The meanings of this term adorn the book of Sugar.
It is worse if we cannot understand from the context what concept we are talking about, as, for example, occurs with the term construction, which I examined in the last article, The Concept of System and Construction. Their place in the design of information systems . The problem is that it is not clear what is meant now: an object, or a set of objects, because both are denoted by one name: a construction. For example, if we say that a platoon has been dug in, it is clear that in this context we are talking about a platoon system, which we consider as an object. If we say that a platoon consists of 20 fighters, then we are talking about a platoon, which we regard as a set of objects. At the same time, the first meaning of the term “platoon” is in no way connected with the second meaning of the same word “platoon”. However, if you do not specify the context, it is impossible to understand what is being discussed now: about the set, or about the object? As our friend philologist said, the topic is well-known, and it did not happen by chance. In order not to overload the language with unnecessary terms, some of them may denote different concepts. However, when formalizing statements, we encounter difficulties in translation, because here we are forced to distinguish them explicitly.
Even worse, when the concepts are two, and the definition is one, as happened with the term system. System engineers have taken away one of the definitions of the term system, and I do not know how they now distinguish them. And it’s very bad when there are a lot of concepts, and the definition is one, like the term process. This is a function, and a functional structure, and a class of functions, and a type of functions, and an operation, and a sequence of operations, and a typical sequence of operations, and a class of sequences of operations, and a class of operations, and so on. As a result, almost everyone involved in the process approach cannot distinguish function from a scenario, cannot say that they model using IDEF0 notation, how models in IDEF0 notation differ from models in BPMN notation, and models in BPMN notation from Gantt diagrams . Therefore, every time I hear this term, I get very tense, because in order to translate it, I have to simultaneously hold in my mind all the many possible concepts in order to understand from the context what the interlocutor is talking about now. Very often, the interlocutor uses the term process in different ways, sometimes in the same sentence. Understand the interlocutor is impossible, and the conversation can be stopped. There is another mistake when the term process denoting a function contrasts the term with a function denoting a function. And they literally say the following: see, function is bad, and the process (in the sense of function) is good. In fact, the manipulation of the “invention” of the process approach to management is based on this balance of terms. I will tell you about how this looks from the side in a separate article, because now, after defining the term construction, we have everything necessary for this.
In this article I will consider only a small fragment concerning the definition of the term “process” in that part of it, which overlaps with the current topic of conversation: the constructions and the rules for submitting definitions.
Let me remind you that we decided to call the process a set of sequential actions aimed at achieving a certain result. The combination, composition and so on. - this is the designation of the set. That is, in this definition, a process is defined as a set of objects - actions. The term action is not simple, but we still leave it aside. For now, let's look at the following thesis: a lot of actions by someone are aimed at achieving a result. It is clear that the set can not have this property. A set can have properties: the number of elements in a set, the average duration of actions included in a given set, and so on. (This is exactly what we saw in the definition of the term train. This is a set of cars driven by a locomotive. It is clear that a set cannot move. Elements of this set can move, or an object synthesized from this set. When we say that a train moves we mean that the object is moving, not the set). In the process definition, only the object synthesized from the sequence of actions can be directed to the result. Consequently, when it comes to directing an object to a result, the process was meant as an object. It turns out that the process as an object is not defined, but there is a link to it. Do you know the problem with the definition of the system? The terms system, design, process simultaneously designate both a set of objects and an object. If these concepts are divorced in the design definition, then in the system definition, and now, as we see, and the process - they are merged in one bottle. But do not blame only on these definitions. All definitions in dictionaries are wrong. They give only one definition of the term, but not another. For example, when they say that an airplane is a vehicle, they do not say that an airplane is a multitude {tail, fuselage, wings, landing gear, engines}. When they say that a train is a multitude of cars, they do not say that a train is a vehicle. Our familiar philologist confirmed that this problem is known in lexicography. If she is known to philologists, then why don't we know anything about her? And how does our ignorance affect how we build models?
Synthesis of an airplane as a set generates an airplane as an object. The synthesis of a train as a set of cars generates a train as an object. Synthesis of a sequence of actions generates what? What is a process as an object? For myself, I decided that this action. That is, as the functions in the synthesis generate functions, so the actions in the synthesis generate actions. By analogy with the construction process defines two different concepts:
Since any action can be divided into actions, the second definition can be simplified. It turns out:
A process is an action directed by someone to achieve a certain result.
We got the same problem as with the definition of the term system. As we found out, all objects are systems (o), and now we understand that all actions are processes (o). That is why I would not call an object a system, but an action a process. I would say that there are objects and their designs. Then we could give different definitions of the aircraft and its design, the train and its design clearly.
Having dealt with the first task, let's move on to the second. We need to define what it means to be “aimed at achieving a certain result”? That is, the definition clearly has a subjective point of view. Someone should tell us whether this action is aimed at achieving a result, or not? In any definition there is a link to the subject. For example, a swimming facility is an object on which to swim. But who decides whether to swim at the facility? Some kind of subject. As long as everyone agrees that you can swim at this facility, we can refer it to swimming facilities. But as soon as a subject appears who says that this cannot be done, we will have a problem: can this object be called a swimming means, or not? If in the swimming facility we can still agree and the cases when the boat is not recognized as a floating craft can be counted on one hand, then it’s not so simple with the actions. You can often hear that the action, from the side of one project stakeholder is considered useful, but from the point of view of the other - not. If the action is useless, then it can not be called a process. It turns out that the action in one IP can be at the same time a process and not be it. Nothing wrong with that, but imagine a situation where the customer knows that the action is meaningless, harmful, but he asks us to automate it. Maybe he is an idiot? I do not know. But how do we call what he asks us to automate? The process is no longer possible. What is the name of the action, not directed anywhere by anyone?
It's time to remember the book Sugar. He writes that in order to define a new concept we must:
An example of a good definition is the definition of a material point given in physics. A material point is a material object whose size within the framework of the problem to be solved can be neglected.
In addition, since the same subject can be classified in different ways, a complete definition of the concept should include the most detailed description of each such superclass, and differential properties. For example, a seaplane can be interpreted as a subclass of aircraft and a subclass of craft. In the definition of a seaplane, you must specify both these superclasses and explain the differential properties of seaplanes that distinguish them from airplanes and boats.
Let us return to the definition of the process (o). Let me remind you that the Process is an action aimed at achieving a result by someone.
What confuses me in this definition? What gesheft is not clear about the constraint imposed on the action: focus on the result? What class of problems is solved more easily if we limit ourselves to this rather tight restriction? The answer is no! That is, there is a restriction, but gesheft is not. Senseless and harmful restriction. It is to this thesis that I let you down all this time. There are definitions that give us nothing but problems. They are meaningless and counterproductive. It makes no sense to impose a restriction if it does not help us solve our problems.
Now, about the emergence of systems. Does anyone know what is gesheft from this restriction? What class of problems is solved more simply if the system is recognized as emergent?
What gesheft we have from the fact that we limit the number of objects in the construction of more than one? why it is impossible to consider constructions with zero number of elements, or with one element?
Since, as analysts, we have to give definitions all the time, it was nice to remember these rules.
Why am I so seriously considering the topic of definitions? Because I regularly meet definitions that cannot be used in practice. And I do not see a trend towards improvement. To begin with, we will analyze the term process by analogy with the way we have discussed the term system in the last article.
I have seen many process definitions, each of which contradicts the other. Here is one of them:
A process is a set of consecutive actions aimed at achieving a certain result.
')
The reason for the contradictions is that with one term “process” means a lot of different concepts (I counted up to 30 at one time) and the definitions are trying to give him only one. If they were all listed separately, there would be 30 definitions of different concepts, denoted by one word.
Well, if the word has only a few meanings that are easy to read from the context. For example, the meaning of the term "spit" is easily guessed from the context. The meanings of this term adorn the book of Sugar.
It is worse if we cannot understand from the context what concept we are talking about, as, for example, occurs with the term construction, which I examined in the last article, The Concept of System and Construction. Their place in the design of information systems . The problem is that it is not clear what is meant now: an object, or a set of objects, because both are denoted by one name: a construction. For example, if we say that a platoon has been dug in, it is clear that in this context we are talking about a platoon system, which we consider as an object. If we say that a platoon consists of 20 fighters, then we are talking about a platoon, which we regard as a set of objects. At the same time, the first meaning of the term “platoon” is in no way connected with the second meaning of the same word “platoon”. However, if you do not specify the context, it is impossible to understand what is being discussed now: about the set, or about the object? As our friend philologist said, the topic is well-known, and it did not happen by chance. In order not to overload the language with unnecessary terms, some of them may denote different concepts. However, when formalizing statements, we encounter difficulties in translation, because here we are forced to distinguish them explicitly.
Even worse, when the concepts are two, and the definition is one, as happened with the term system. System engineers have taken away one of the definitions of the term system, and I do not know how they now distinguish them. And it’s very bad when there are a lot of concepts, and the definition is one, like the term process. This is a function, and a functional structure, and a class of functions, and a type of functions, and an operation, and a sequence of operations, and a typical sequence of operations, and a class of sequences of operations, and a class of operations, and so on. As a result, almost everyone involved in the process approach cannot distinguish function from a scenario, cannot say that they model using IDEF0 notation, how models in IDEF0 notation differ from models in BPMN notation, and models in BPMN notation from Gantt diagrams . Therefore, every time I hear this term, I get very tense, because in order to translate it, I have to simultaneously hold in my mind all the many possible concepts in order to understand from the context what the interlocutor is talking about now. Very often, the interlocutor uses the term process in different ways, sometimes in the same sentence. Understand the interlocutor is impossible, and the conversation can be stopped. There is another mistake when the term process denoting a function contrasts the term with a function denoting a function. And they literally say the following: see, function is bad, and the process (in the sense of function) is good. In fact, the manipulation of the “invention” of the process approach to management is based on this balance of terms. I will tell you about how this looks from the side in a separate article, because now, after defining the term construction, we have everything necessary for this.
In this article I will consider only a small fragment concerning the definition of the term “process” in that part of it, which overlaps with the current topic of conversation: the constructions and the rules for submitting definitions.
Let me remind you that we decided to call the process a set of sequential actions aimed at achieving a certain result. The combination, composition and so on. - this is the designation of the set. That is, in this definition, a process is defined as a set of objects - actions. The term action is not simple, but we still leave it aside. For now, let's look at the following thesis: a lot of actions by someone are aimed at achieving a result. It is clear that the set can not have this property. A set can have properties: the number of elements in a set, the average duration of actions included in a given set, and so on. (This is exactly what we saw in the definition of the term train. This is a set of cars driven by a locomotive. It is clear that a set cannot move. Elements of this set can move, or an object synthesized from this set. When we say that a train moves we mean that the object is moving, not the set). In the process definition, only the object synthesized from the sequence of actions can be directed to the result. Consequently, when it comes to directing an object to a result, the process was meant as an object. It turns out that the process as an object is not defined, but there is a link to it. Do you know the problem with the definition of the system? The terms system, design, process simultaneously designate both a set of objects and an object. If these concepts are divorced in the design definition, then in the system definition, and now, as we see, and the process - they are merged in one bottle. But do not blame only on these definitions. All definitions in dictionaries are wrong. They give only one definition of the term, but not another. For example, when they say that an airplane is a vehicle, they do not say that an airplane is a multitude {tail, fuselage, wings, landing gear, engines}. When they say that a train is a multitude of cars, they do not say that a train is a vehicle. Our familiar philologist confirmed that this problem is known in lexicography. If she is known to philologists, then why don't we know anything about her? And how does our ignorance affect how we build models?
Synthesis of an airplane as a set generates an airplane as an object. The synthesis of a train as a set of cars generates a train as an object. Synthesis of a sequence of actions generates what? What is a process as an object? For myself, I decided that this action. That is, as the functions in the synthesis generate functions, so the actions in the synthesis generate actions. By analogy with the construction process defines two different concepts:
- A process is a set of sequential actions, the synthesis of which forms an action directed by someone to achieve a certain result.
- A process is an action directed by someone to achieve a certain result, the design of which can be represented as a set of actions.
Since any action can be divided into actions, the second definition can be simplified. It turns out:
A process is an action directed by someone to achieve a certain result.
We got the same problem as with the definition of the term system. As we found out, all objects are systems (o), and now we understand that all actions are processes (o). That is why I would not call an object a system, but an action a process. I would say that there are objects and their designs. Then we could give different definitions of the aircraft and its design, the train and its design clearly.
Having dealt with the first task, let's move on to the second. We need to define what it means to be “aimed at achieving a certain result”? That is, the definition clearly has a subjective point of view. Someone should tell us whether this action is aimed at achieving a result, or not? In any definition there is a link to the subject. For example, a swimming facility is an object on which to swim. But who decides whether to swim at the facility? Some kind of subject. As long as everyone agrees that you can swim at this facility, we can refer it to swimming facilities. But as soon as a subject appears who says that this cannot be done, we will have a problem: can this object be called a swimming means, or not? If in the swimming facility we can still agree and the cases when the boat is not recognized as a floating craft can be counted on one hand, then it’s not so simple with the actions. You can often hear that the action, from the side of one project stakeholder is considered useful, but from the point of view of the other - not. If the action is useless, then it can not be called a process. It turns out that the action in one IP can be at the same time a process and not be it. Nothing wrong with that, but imagine a situation where the customer knows that the action is meaningless, harmful, but he asks us to automate it. Maybe he is an idiot? I do not know. But how do we call what he asks us to automate? The process is no longer possible. What is the name of the action, not directed anywhere by anyone?
It's time to remember the book Sugar. He writes that in order to define a new concept we must:
- Name the superclass of objects. For example, we say that a chair is furniture. Furniture is a wider class of objects than a class of chairs.
- Identify the distinctive features (differential properties) of objects of this class from objects of a more general class. For example, a chair is an object that has four legs and a back.
- Describe the class of tasks, functions performed, and so on. Solved (performed) by these objects: the chair is designed for sitting. These tasks should explain why the previously mentioned differential features were singled out: 4 legs — for stability (and why not three?) Back — so as not to strain your back when sitting on a chair, and so on.
- Describe the restrictions that impose differential features on the tasks to be performed, the functions performed. For example, you cannot keep things in a chair.
An example of a good definition is the definition of a material point given in physics. A material point is a material object whose size within the framework of the problem to be solved can be neglected.
- The superclass is material objects.
- Differential properties: from the point of view of the subject, the size of the material body is zero, unlike other bodies.
- The task is to calculate the trajectory of the material body. The reason for the differential property is to simplify the calculations.
- I will not enumerate restrictions, because in good textbooks of physics a lot of time is devoted to the question of when one can imagine a material body as a point, and when it is not.
In addition, since the same subject can be classified in different ways, a complete definition of the concept should include the most detailed description of each such superclass, and differential properties. For example, a seaplane can be interpreted as a subclass of aircraft and a subclass of craft. In the definition of a seaplane, you must specify both these superclasses and explain the differential properties of seaplanes that distinguish them from airplanes and boats.
Let us return to the definition of the process (o). Let me remind you that the Process is an action aimed at achieving a result by someone.
- Process class - action class
- Differential sign - focus on achieving results.
- The purpose of the introduction of the differential feature is unknown to me. The range of tasks that this feature helps to solve is also not clear.
- Limitations: if two stakeholders do not agree on the usefulness of the action, it cannot be called a process.
What confuses me in this definition? What gesheft is not clear about the constraint imposed on the action: focus on the result? What class of problems is solved more easily if we limit ourselves to this rather tight restriction? The answer is no! That is, there is a restriction, but gesheft is not. Senseless and harmful restriction. It is to this thesis that I let you down all this time. There are definitions that give us nothing but problems. They are meaningless and counterproductive. It makes no sense to impose a restriction if it does not help us solve our problems.
Now, about the emergence of systems. Does anyone know what is gesheft from this restriction? What class of problems is solved more simply if the system is recognized as emergent?
What gesheft we have from the fact that we limit the number of objects in the construction of more than one? why it is impossible to consider constructions with zero number of elements, or with one element?
Source: https://habr.com/ru/post/328228/
All Articles