Society as a network
(Continuation of the reflections begun in the article Factors of the social realization of users ).
From an interview with Arkady Volozh, CEO and co-owner of Yandex:
“Services that repeat the basic life processes work well on the Internet: information search, communication, social networks, advertising - all this exists in real life. Internet technologies simply transfer these processes to another level. Our strategy is to create what is in real life and can be enhanced by the Internet. ”
The term “real life” here, apparently, should be understood mainly as a traditional society, a social environment. Let's see what society can be in projection on the Internet space. By the way, why is the Internet called a network? Probably, by tradition - originally it was a network of wired computers. But this is the same as if by computer we mean only “iron”. The Internet today is hard + soft, with soft programming people, their activity. The Internet is a peculiar reflection of traditional society and we can see in this reflection that society itself is also a network. It includes not only people, the objects of this network are heterogeneous and connected by heterogeneous relations. For example, a person may be a member of communities and organizations (ownership), within organizations they are in a boss-subordinate relationship, while writing a book, planting a tree, having a child and building a house - they will be linked to a person with a cause-effect relationship. He also owns many things (property relations). These things themselves belong to classifications according to all sorts of other criteria, for example, the book relates to the subject, genre, national tradition, etc. (in other words, the object “book” is related to the relation of belonging with the objects “theme”, “genre”, etc.). In turn, the “theme” is associated with many other topics, materials, books, authors, and organizations. From these arguments, a model of the network “objects + connections” is viewed, which in its abstractness is simply a definition of a graph . It is suitable as a basis for modeling society as a resource environment and at the same time is simple enough to be implemented within a separate service in which users create content objects of different types and link them with different types of links.
')
In general, I have not heard that such a model was considered one of the ways to streamline content. The main methods of classification usually include hierarchical structures (trees, subfolders, etc.) and the method of faceted classification. In the field of artificial intelligence, types of the model “objects + connections” (semantic networks, frame systems) are considered to be a way of representing knowledge . In fact, the topic of ontologies in computer science also applies to this. Perhaps the reason why these ideas are not counted as classification methods lies in their orientation towards machine use, whereas people usually call classifications what is more focused on people's perceptions. Or maybe the point is in the discrepancy between the concepts of ordering and classification - the second always implies the first, but the first is not always associated with the second. In any case, the “objects + communications” network is an organized environment and can be considered a way of streamlining.
If the types of content objects and the types of relationships between them are such that they allow you to implement the main traditional types of Internet user activity, then parallel to this activity will be the ordering of content. Since links between objects carry some information about these objects, these links are actually a kind of metadata. This approach can be called user-generated metacontent . Another type of metadata includes tags, and a similar name is invented for them - people-powered metadata . There is also a third, concise definition - Folksonomy . All of them reflect the meaning of the idea of ​​“popular content ordering”, but do not reflect the specifics of the ordering methods.
The peculiarity of the “objects + connections” model is that the same object can belong simultaneously to different classifications, structures, sets. In the hierarchical (tree-like) method of classification, this is not the case, in the facet one, but for me this method is rather complicated and I still have no idea how it can be implemented in an Internet-project that everyone understands. Meanwhile, the mentioned feature is key in terms of solving the problems formulated in my previous article, Factors of the social realization of users . Consider the points from it:
1. In the network “objects + links” content objects are positioned approximately as in Wikipedia, i.e. belong to the common “ocean”, in which there is no “spot of universal temporary attention,” to which all materials (authors) aspire due to the rating. By the way, the idea of semantic Wikipedia with its typed links almost exactly reproduces the model “objects + links” described here. As far as I understand, there is no typing of created objects as superfluous as all of them are assigned the only type “article for encyclopedia” by default. Actually, the principle of wiki on Wikipedia is embodied by this very type, since it contains the positioning of the network encyclopedia as something jointly created and useful for all. However, if you allow more types of content objects, as well as introduce a system of user polls, then such an environment can be customized for other wiki-positioning or wiki-activity options. In connection with the "common ocean" I will note a particular point related to commenting, since this is one of the most common types of activity on the Internet. The problem is that in traditional services the comments do not have the same “rights” and possibilities as the author's commented text, although they may be no less informative and interesting. In the network "objects + communication" all objects are equal; content positioning, the context for it is determined by its author with the help of object types and links.
Items 2 and 3 are more technical, they relate not so much to the advantages of the “objects + connections” model (although this is a matter for separate consideration), but rather to ordered content in general. But in this case, the mere fact that there is an order in the “objects + connections” network is enough.
4. Combining chaos and order - how to combine the unpredictable nature of the discussions with the resulting orderly knowledge? From the point of view of the “objects + connections” model, the discussion is a set of objects, roughly speaking, of text type, connected by a kind of causal link, let's call it conditionally “source-response”. But nothing prevents these objects from having connections (to belong) to any other structures, including hierarchical tree structures. As material is comprehended, these connections can be established afterwards.
The division of related topics and the association of distant ones - in fact, this is also the creation of additional classifications to existing ones, which is ensured by the model under consideration.
Qualitative transformations of content - they need “one place” (one service, although, by the way, it is not so necessary - you can think of some nadservice features) and simply ordered content by subject. We already have it. But we can also add that in itself the detailed typification of content objects and their belonging simultaneously to many hierarchies (systematization according to many different signs and criteria) actually means a high degree of understanding and study of the material.
As the predecessors of this approach, you can specify the non-network project TheBrain , in which different types of objects can be connected by different types of connections. The difference between the proposed approach is a) orientation towards the Internet, b) wiki ideologies, c) other types of links more suitable for Internet activity. Other predecessors can be considered the above-mentioned ontologies and those created by people, but machine-oriented knowledge bases of expert systems; in this case, the differences are related to a) the Internet, b) people, c) the representation of society, and not knowledge. The last consideration is interesting in that it means a different understanding of the concept of the Semantic Web : the future of the Internet is associated with identifying the potential of society as an organized resource environment in which people play the main role and their comprehension activities, and the results of their activity (in particular, in the form of metadata) are used by cars. This corresponds to the idea of ​​the dominance of the creative component over the routine; nevertheless, we should rather talk about the synergistic effect of the interaction of society and machines. In fact, the latter also belong to the common network, i.e. are one of the types of its objects and represent a “computational” resource. (Udt 04/06/2009 This can also be expressed by the thesis that it is easier to identify and use the potential of people than to try to teach machines to do what people can do. The last thing you really need to do is to reveal and push the limits of the capabilities of machines and their area of ​​efficiency, but apparently, there will always be an area where people are still more effective. Understanding this determines both effective development strategies and the “human” area of ​​the Internet, and the machine, and the interaction of these areas. )
As stated at the beginning of the article, startups essentially program the activity of people in the network. Perhaps, the proposed reflections can be attributed to the creation of a certain language or toolkit of this kind of programming, but focused not on startups, but on ordinary users of the network who in such activity could find great opportunities for their own social realization. For example, if you create complex objects from simple objects, which will also belong to a common network and have connections with both their internal elements and external ones, this will give users the ability to very detailed and accurate reflection of various aspects of reality. These last statements, however, are rather vague; they require further detailing of the material, which not all could be performed in this article due to the expansion of its volume; it also requires further reflection and discussion.
From an interview with Arkady Volozh, CEO and co-owner of Yandex:
“Services that repeat the basic life processes work well on the Internet: information search, communication, social networks, advertising - all this exists in real life. Internet technologies simply transfer these processes to another level. Our strategy is to create what is in real life and can be enhanced by the Internet. ”
The term “real life” here, apparently, should be understood mainly as a traditional society, a social environment. Let's see what society can be in projection on the Internet space. By the way, why is the Internet called a network? Probably, by tradition - originally it was a network of wired computers. But this is the same as if by computer we mean only “iron”. The Internet today is hard + soft, with soft programming people, their activity. The Internet is a peculiar reflection of traditional society and we can see in this reflection that society itself is also a network. It includes not only people, the objects of this network are heterogeneous and connected by heterogeneous relations. For example, a person may be a member of communities and organizations (ownership), within organizations they are in a boss-subordinate relationship, while writing a book, planting a tree, having a child and building a house - they will be linked to a person with a cause-effect relationship. He also owns many things (property relations). These things themselves belong to classifications according to all sorts of other criteria, for example, the book relates to the subject, genre, national tradition, etc. (in other words, the object “book” is related to the relation of belonging with the objects “theme”, “genre”, etc.). In turn, the “theme” is associated with many other topics, materials, books, authors, and organizations. From these arguments, a model of the network “objects + connections” is viewed, which in its abstractness is simply a definition of a graph . It is suitable as a basis for modeling society as a resource environment and at the same time is simple enough to be implemented within a separate service in which users create content objects of different types and link them with different types of links.
')
In general, I have not heard that such a model was considered one of the ways to streamline content. The main methods of classification usually include hierarchical structures (trees, subfolders, etc.) and the method of faceted classification. In the field of artificial intelligence, types of the model “objects + connections” (semantic networks, frame systems) are considered to be a way of representing knowledge . In fact, the topic of ontologies in computer science also applies to this. Perhaps the reason why these ideas are not counted as classification methods lies in their orientation towards machine use, whereas people usually call classifications what is more focused on people's perceptions. Or maybe the point is in the discrepancy between the concepts of ordering and classification - the second always implies the first, but the first is not always associated with the second. In any case, the “objects + communications” network is an organized environment and can be considered a way of streamlining.
If the types of content objects and the types of relationships between them are such that they allow you to implement the main traditional types of Internet user activity, then parallel to this activity will be the ordering of content. Since links between objects carry some information about these objects, these links are actually a kind of metadata. This approach can be called user-generated metacontent . Another type of metadata includes tags, and a similar name is invented for them - people-powered metadata . There is also a third, concise definition - Folksonomy . All of them reflect the meaning of the idea of ​​“popular content ordering”, but do not reflect the specifics of the ordering methods.
The peculiarity of the “objects + connections” model is that the same object can belong simultaneously to different classifications, structures, sets. In the hierarchical (tree-like) method of classification, this is not the case, in the facet one, but for me this method is rather complicated and I still have no idea how it can be implemented in an Internet-project that everyone understands. Meanwhile, the mentioned feature is key in terms of solving the problems formulated in my previous article, Factors of the social realization of users . Consider the points from it:
1. In the network “objects + links” content objects are positioned approximately as in Wikipedia, i.e. belong to the common “ocean”, in which there is no “spot of universal temporary attention,” to which all materials (authors) aspire due to the rating. By the way, the idea of semantic Wikipedia with its typed links almost exactly reproduces the model “objects + links” described here. As far as I understand, there is no typing of created objects as superfluous as all of them are assigned the only type “article for encyclopedia” by default. Actually, the principle of wiki on Wikipedia is embodied by this very type, since it contains the positioning of the network encyclopedia as something jointly created and useful for all. However, if you allow more types of content objects, as well as introduce a system of user polls, then such an environment can be customized for other wiki-positioning or wiki-activity options. In connection with the "common ocean" I will note a particular point related to commenting, since this is one of the most common types of activity on the Internet. The problem is that in traditional services the comments do not have the same “rights” and possibilities as the author's commented text, although they may be no less informative and interesting. In the network "objects + communication" all objects are equal; content positioning, the context for it is determined by its author with the help of object types and links.
Items 2 and 3 are more technical, they relate not so much to the advantages of the “objects + connections” model (although this is a matter for separate consideration), but rather to ordered content in general. But in this case, the mere fact that there is an order in the “objects + connections” network is enough.
4. Combining chaos and order - how to combine the unpredictable nature of the discussions with the resulting orderly knowledge? From the point of view of the “objects + connections” model, the discussion is a set of objects, roughly speaking, of text type, connected by a kind of causal link, let's call it conditionally “source-response”. But nothing prevents these objects from having connections (to belong) to any other structures, including hierarchical tree structures. As material is comprehended, these connections can be established afterwards.
The division of related topics and the association of distant ones - in fact, this is also the creation of additional classifications to existing ones, which is ensured by the model under consideration.
Qualitative transformations of content - they need “one place” (one service, although, by the way, it is not so necessary - you can think of some nadservice features) and simply ordered content by subject. We already have it. But we can also add that in itself the detailed typification of content objects and their belonging simultaneously to many hierarchies (systematization according to many different signs and criteria) actually means a high degree of understanding and study of the material.
As the predecessors of this approach, you can specify the non-network project TheBrain , in which different types of objects can be connected by different types of connections. The difference between the proposed approach is a) orientation towards the Internet, b) wiki ideologies, c) other types of links more suitable for Internet activity. Other predecessors can be considered the above-mentioned ontologies and those created by people, but machine-oriented knowledge bases of expert systems; in this case, the differences are related to a) the Internet, b) people, c) the representation of society, and not knowledge. The last consideration is interesting in that it means a different understanding of the concept of the Semantic Web : the future of the Internet is associated with identifying the potential of society as an organized resource environment in which people play the main role and their comprehension activities, and the results of their activity (in particular, in the form of metadata) are used by cars. This corresponds to the idea of ​​the dominance of the creative component over the routine; nevertheless, we should rather talk about the synergistic effect of the interaction of society and machines. In fact, the latter also belong to the common network, i.e. are one of the types of its objects and represent a “computational” resource. (Udt 04/06/2009 This can also be expressed by the thesis that it is easier to identify and use the potential of people than to try to teach machines to do what people can do. The last thing you really need to do is to reveal and push the limits of the capabilities of machines and their area of ​​efficiency, but apparently, there will always be an area where people are still more effective. Understanding this determines both effective development strategies and the “human” area of ​​the Internet, and the machine, and the interaction of these areas. )
As stated at the beginning of the article, startups essentially program the activity of people in the network. Perhaps, the proposed reflections can be attributed to the creation of a certain language or toolkit of this kind of programming, but focused not on startups, but on ordinary users of the network who in such activity could find great opportunities for their own social realization. For example, if you create complex objects from simple objects, which will also belong to a common network and have connections with both their internal elements and external ones, this will give users the ability to very detailed and accurate reflection of various aspects of reality. These last statements, however, are rather vague; they require further detailing of the material, which not all could be performed in this article due to the expansion of its volume; it also requires further reflection and discussion.
Source: https://habr.com/ru/post/55728/
All Articles