Some explanations:
In May 2012, I was visited by the idea of systematization of the process of writing technical texts (mainly reports and articles). Taking as a basis some ergonomic assumptions, I introduced simple assumptions and made an attempt to create a text “on the fly” that would describe itself as a method of constructing texts. I give the result of such an attempt.
Assumptions:
- Suppose there is a certain “hyper volume of consciousness” that can simultaneously operate with several units of information, which we conditionally call “lexemes”. Let us denote this phenomenon as “the operational memory of the brain”.
- Suppose the number of tokens that can be simultaneously placed in the RAM of the brain can not exceed 5..7 units.
- Suppose the order of receipt of tokens in the operational memory does not matter for consciousness. (By analogy with the phenomenon: “According to rzelulattas, or aseudoadic one long nonglissokgo univertiset, do not have time ..., etc.”)
- Suppose the presence of “extra” lexemes in the operational memory is not a problem from the point of view of their processing by consciousness.
- Suppose there is no significant forgetting and transformation of tokens.
Then you can create a method for constructing texts, such that the text obtained as a result of its application will be well structured, devoid of uncontrollable redundancy, will not have unrelated elements, and can also be easily represented as a graph. In other words, this text will be readable, “understandable” and “easy-to-remember”.
Chapter 1
1. Within each paragraph, the logic of the following material does not matter, because each paragraph is a unit that the operational memory of the brain is guaranteed to keep. A paragraph is a subsystem within a supersystem — the lower level chapter.
2. Each paragraph must be formulated in such a way that it is obvious what connections it has with which paragraphs. If in each paragraph there is a certain “main idea” (“core”) - from these “basic thoughts” there can be, in turn, a cell “unit” of information that can be guaranteed to hold the operational memory.

Fig.1 Paragraph of the cores.
')
3. Thus, the size of the chapter is limited: the chapter should consist of so many paragraphs that one paragraph can be made up of their cores. The question remains - is it possible to do the same with the heads? Is it possible to select a core inside each chapter and limit the number of chapters to the possibility of constructing a paragraph from the kernels of chapters?
4. Paragraphs within the text and chapters should be followed in such a way that the logic of presentation of the material has a unidirectional character. This rule, of course, is a recommendation. Probably should be attributed not to the paragraph as a whole, but to the cores of paragraphs. Paragraph kernels should follow the unidirectional rule. Violation of this rule leads to references to kernels that do not exist for the reader, which contradicts the principle “paragraph from kernels”.
5. One of the problems associated with the compilation of texts is the limited possibilities of the human brain. One way or another, there comes the limit of learning. How quickly this boundary comes depends on two factors. The first is the transparency of the text, the second is its clarity. Moreover, transparency depends on the author of the text, and clarity depends on the reader.
6. A phenomenon that violates the transparency of the text, among other things, are edits. Edits, both within the paragraph, and within the chapters of all levels. Edits cause intermixing and mismatching of paragraphs, leading to a violation of the rule of one-pointedness of following the core of paragraphs. An important question in the compilation of texts is the question - how to edit the text so that the negative effect of them is insignificant?
7. Perhaps, forming paragraph chapters should follow the mnemonic rule 5..7. This rule is derived empirically (for the US Navy) and is that the brain's operational memory can simultaneously operate with a number of objects approximately equal to 5..7. Following this principle should increase the transparency of the text.
8. This chapter has 8 paragraphs (including this one), an attempt will now be made to single out the core of paragraphs and make a link diagram and a text paragraph out of them.
Chapter 2
9. So, summarizing paragraphs 1..7 in accordance with the communication diagram (Fig. 2), you can make a paragraph with the following content: “Limiting the capabilities of the brain leads to the need to construct texts so that they are as convenient as possible for perception. For the convenience of perception, the logic of the text should be unidirectional, the amount of information in each paragraph and in each chapter should
not be
too large . Moreover, this information should be as transparent as possible to achieve, as a result, maximum clarity on the part of the reader. ”

Fig.2 Paragraph link diagram 1..7. Direct references of paragraph 9 are indicated by a normal line, indirect links are indicated by a dotted line.
10. Apparently, paragraph 9 directly refers to paragraphs 1, 4, 5, indirectly to 2, 3, 7, and does not refer at all to 6. This means, on the one hand, that the author of the text considers 6 as a secondary paragraph and, on the other hand, , denotes the insufficient experience of the author in drawing up capacious and at the same time not too bulky paragraphs.
11. Following the tradition of presenting texts, this text (consisting of one chapter) should be divided into an introduction, a main part and a conclusion. This can be achieved only with the help of edits: move paragraph 9 to the first place or, on its basis, create a new paragraph transmitting similar information in a more general and concise form. Obviously one thing: within the chapter should be left no more than 7 paragraphs, and all other paragraphs to make a separate chapter.
12. Also, the name of the text as a whole should be stated. The name should be the “core kernel of chapters”. One of the ways of formulating the name can be the selection of the core of the summarizing paragraph 9 and the introduction of the information revealing the name in the introduction. Variants of titles: “Application of the principle of transparency to the formation of texts” or “Method of constructing texts in accordance with the principle of transparency”, or “Method of constructing texts that are easy to read”. The last name is most relevant, because does not violate the principles of unidirectional logic of the text.
13. Next, a diagram of the connections of paragraphs 1..12 will be compiled, a paragraph from the kernels 10..12 is formulated, an introduction and a conclusion of the text are compiled. References will not be given, because it has only one name (source of rule 5..7). This source is the ergonomics handbook “Human engineering guide to equipment design” 1971.

Diagram 3 Paragraph Relationship Chart 1..12.
14. Paragraph from cores 10, 11, 12: The logic of links between paragraphs should be tracked. The rule of one-pointedness of the logic of the sequence of nuclei must be observed in a separate text at all its levels. Convenient text should have a simple transparent title.
15. To compile the introduction and conclusion, in the framework of the proposed method, based on the hierarchy of the structure of the text, it is not necessary to compile summarizing paragraphs (such as 9 and 14) to re-study the entire text from beginning to end. It is enough to select the cores 9 and 14 and make two review paragraphs of them. These two paragraphs, following the traditions of drafting texts, should state the same thing, but with some differences. Namely: the introduction should be focused on previous research on the topic under consideration, and the conclusion - on future ones.
16. Introduction:
In the process of creating texts, regardless of their scope, there is a problem associated with their proliferation and, as a result, a violation of the internal logic and ease of assimilation by the reader. To prevent such phenomena, it is necessary to formulate an easy-to-use method that allows you to control the logic of the text in such a way that the control of this logic does not depend on the size of the text being compiled. Analysis of existing methods of this kind revealed the problem of complete disregard of the adaptability of the method to the expansion of the text. The author made an attempt to build a convenient method for the formation of texts, since a readable text can be composed solely as a result of the operation of a method that is convenient for use.
17. Conclusion:
As a result of the work on the creation of a text compilation method, an algorithm was developed aimed at preserving the stability of the control of the logic of texts regardless of the growth of their volume. The method, in its current edition, allows you to select in the text a hierarchy of paragraphs, to select among paragraphs summarizing, gives an approach to improving the transparency of the text, taking into account the psychology of text perception. This text is built in the process of forming the method. Due to this recursiveness, it cannot serve as a benchmark for the application of the method. However, recursiveness has positive qualities - it shows the process of forming the method “as it is” - reveals the chronology of the construction of paragraphs, the evolution of the method in the process of its creation, etc. Probably, a valuable contribution to the development of the method can be made from the standpoint of hermeneutics.
Comments:
The introduction turned out to be at the end (paragraph 16); this is connected with the chronology of the writing of the text and does not bear the basic load. Practice has shown that the construction of texts that strictly correspond to the method takes an unreasonably long time, however, with the method in mind, it is possible, at least, not to bring confusion, where possible. The application of the method to bulky reports of the “one copy - one pack of paper” class showed its low sensitivity to the spread of texts. "The problem of edits" is removed timely transition to new versions of the text.
In conclusion, I would like to express my appreciation to Ana H. for beer, pistachios and valuable editing.