Thematic content generation on the Internet. A look into the future
Fundamental conflict
The struggle of Internet resources for users (for authors of content and audience of readers) leads to the fact that the author's content does not get into the field of attention of the maximum amount of its target audience. Authors can not be active at once at all sites and are forced to focus on certain ones.
Separation of content from services
A step towards resolving the conflict may be the creation of an “Internet of content”. Its basis is not a web page, which is a container for content, but content units themselves. We assign a unique address to each such unit and supply it with certain metadata — creation time, author (s), subject tags, probably something else.
This is in line with the ideas of the Semantic Web, except that the latter was conceived machine-oriented, at convenient processing of content by machines, more accurate extraction of meanings, while the proposed approach focuses on people, their perception, convenience and ability to build services on this basis.
Internet content does not cancel sites, it can be built inside the Internet sites. We just need additional services. Suppose, while on a particular site, select a piece of text, fill in metadata (or they are filled in automatically, you only edit) and then a third-party service assigns the address to this content unit (and can be stored in its database along with a link to the resource). I am not an expert in the technical part, maybe this is realizable at the browser level?
')
The term “separation of content from services” was coined by amilner , now I don’t remember exactly where. But it seems the author meant something of his own.
Content positioning
One of the ideas that I carry out and consider important if we are people-oriented is the positioning of the mentioned content units. So to say, the markup of the text on positioning: we divide it into components - “thesis”, “argument”, “opinion”, “hypothesis”, “question”, etc. We make each component (even such small forms) with an independent unit and Positioning is explicitly declared in the metadata.
Firstly, it helps to correctly adjust the perception of the reader, thereby facilitating understanding and communication. We even give different weight to different positioning, for example, the “argument” in the discussion is more important than the “opinion”. In addition, it reflects reality - the thematic division alone does not exhaust speech, positioning is also explicitly or implicitly present.
Attention to small forms
First, it is easier for people to generate content in small forms. Secondly, such forms are more likely to compete for the attention of the audience. Third, they are less subject to change - this is important due to the fact that in the process of development of any topic its structure changes. You can, for example, write an article and at the end understand where to begin. When there are ready-made decorated and meaningful “bricks” and blocks of related considerations, it is easier to build them into various larger logical structures.
Discussion + structuring
People are also easier to generate content in the process of communication, discussions, discussions. A good structuring of the topic is almost equivalent to its good comprehension. It is desirable that these two things are harmoniously interconnected. Otherwise, the mountain of produced content will grow dramatically, but it is not enough to add to the global understanding of the topic, to provoke duplication of topics and walking in a circle.
The question is how to connect discussion and structuring at the interface implementation level. For example, the project Rizzoma (the successor of Google Wave) is trying to do this. In this Google Study, I formulated some criticism of Rizoma.
Top media and search: the principle of uncertainty?
The maximum audience size can be achieved in different ways: if we are talking about a news resource, a tape format, then it provides short-term synchronous attention of a large audience; and if the reader comes from a search query, here we have the asynchronous attention of the audience, which at each moment is not large, but the amount for the entire (indefinite) period of time can be large.
The question is whether it is possible and how to combine these two approaches in one resource.
Separation of rankings from services
In the situation of a large amount of content, its ranking is an important factor determining the attention of the audience. If the content can be separated from the services, then the ranking too. For the “Internet content” such a function looks like a natural analogue of search engines ranking websites on the regular Internet. The content ranking service separated from the sites would use not only the information obtained from the site on which the content was produced. In particular, due to the processing of data on the activity of the author on other sites.
Klout has already implemented the idea of ​​separating the ranking of authors from services, and even displays the author's content, but does not make this service for content. And could - ranking authors and content - things are strongly related. Attitudes toward the author depend on what he produces and does, and this attitude is then transferred to his content — from the author with a high reputation, the content initially receives more attention.
Amount
As suggested above, the user of an arbitrary Internet resource selects the text and supplies it with metadata, these metadata serve as the “address” at which the text falls into the necessary thematic section of the “Internet content” and is correctly positioned there. (Actually, the assignment of another separate identifier - the URI - may not be necessary then).
Here it is necessary to fix an important point - usually people do not like to perform extra actions; tagging the text is just an example of such actions, since the user doesn’t really need it, because the target action — publishing content — will happen without it. Getting into the right section, like on Habré, or to your target audience, like in blogs, happens at the expense of other things.
However, if you create a content object outside the context of a specific service, simply as a thing from the Internet of content that can be used in different services, then providing detailed meta-information is the only way to reach your target audience. The author is vitally interested in this.
So far it turns out that “Internet content” is still a specific service, the minimum function of which is to store data and metadata, which it can transmit to other services through the API.
Further, the same or another service ranks authors and content (I have thoughts about the ranking mechanism, if briefly, this is using the PageRank algorithm in the network of heterogeneous objects — content and people in this case; what is considered a voting link is a separate story, for details see the text on the link).
In this embodiment, it turns out something similar to search engines or Wikipedia, because such a resource does not have a specific thematic positioning. But the issuance of a search query - in my opinion it should be completely different. Not only not the same as search engines, but not even like Wikipedia. More precisely, there is a similarity to Wikipedia - probably an overview article for acquaintance with the topic should be present, but you need an even more or less standard description of other important resources - the top authors most competent in this topic, the list of organizations whose profile this query most closely matches, the list sites. And there is no top content, because The review article itself is top content - and here an important difference from Wikipedia is seen - the content gets to the top as a result of ranking over the entire activity of all users on the Internet, and not individual nameless authors of articles for the wiki encyclopedia. I think a ranked list of references is also needed.
In general, what should be the issue - a separate non-trivial question. How different themes can be standardized by one format of issue and so on. For example, the history of the subject is now a typical section of Wikipedia articles (i.e. some standardization already exists there), and the person Pavel Zakharov claims that he knows the only correct way to tell about the evolution of any subject (in the sense of an algorithm to present it somehow). We also recall the above-mentioned problem of combining the news and search tops - probably there should be a news component in the issue, some ranked tape of new publications (divided by positioning).
In general, the “pyramid structure” seems to be correct, where the top of the pyramid is a review article on the topic, and the lower levels (available as an option) cover an ever-increasing amount of various relevant content.
Of course, the idea to create a “better Wikipedia” may seem ridiculous, but think about it - Wikipedia appeared at the dawn of the Web 2.0 era; In the current era of maturity, it is possible to create a more modern product, taking into account the accumulated experience from related areas and the current trends of the future. Nowadays, only selected altruistic fanatics tend to be the author of Wikipedia, but in its modernized version, everyone would rush there - for the sake of a large amount of relevant audience.
It remains to understand how the concept of combining in a single structuring and discussion space fits into this scheme. Is it worth stimulating discussion right in the “new wikipedia”? And where in this picture of the world is a place for other content services on the Internet. The fundamental problem, which was mentioned at the very beginning, is only on the one hand the problem, and on the other - the incentive to develop and improve content services.
It is only clear that the “New Wikis” themselves may at first be somewhat competing, although in the end one will become strongly dominant - the one that will offer the best ranking and issuing format. As now Google dominates in the field of search, and Facebook in the field of social networks (the current Wikipedia does not count - now it just complements the search). Probably, in the case of the dominance of one of the new wiki, other resources will begin to adapt and aim at integration with it (ie, they will begin to resemble the current applications to social networks). The one who is better able to realize combining structuring with discussions will be able to integrate better and mutually with the giant.
It is also possible that Google or one of the current giants will partially evolve in this direction.
PS I already wrote about almost all of this on Habré, but in different articles and, moreover, now I understood one of my mistakes, which earlier prevented me from more or less consistently presenting the topic.
The struggle of Internet resources for users (for authors of content and audience of readers) leads to the fact that the author's content does not get into the field of attention of the maximum amount of its target audience. Authors can not be active at once at all sites and are forced to focus on certain ones.
Separation of content from services
A step towards resolving the conflict may be the creation of an “Internet of content”. Its basis is not a web page, which is a container for content, but content units themselves. We assign a unique address to each such unit and supply it with certain metadata — creation time, author (s), subject tags, probably something else.
This is in line with the ideas of the Semantic Web, except that the latter was conceived machine-oriented, at convenient processing of content by machines, more accurate extraction of meanings, while the proposed approach focuses on people, their perception, convenience and ability to build services on this basis.
Internet content does not cancel sites, it can be built inside the Internet sites. We just need additional services. Suppose, while on a particular site, select a piece of text, fill in metadata (or they are filled in automatically, you only edit) and then a third-party service assigns the address to this content unit (and can be stored in its database along with a link to the resource). I am not an expert in the technical part, maybe this is realizable at the browser level?
')
The term “separation of content from services” was coined by amilner , now I don’t remember exactly where. But it seems the author meant something of his own.
Content positioning
One of the ideas that I carry out and consider important if we are people-oriented is the positioning of the mentioned content units. So to say, the markup of the text on positioning: we divide it into components - “thesis”, “argument”, “opinion”, “hypothesis”, “question”, etc. We make each component (even such small forms) with an independent unit and Positioning is explicitly declared in the metadata.
Firstly, it helps to correctly adjust the perception of the reader, thereby facilitating understanding and communication. We even give different weight to different positioning, for example, the “argument” in the discussion is more important than the “opinion”. In addition, it reflects reality - the thematic division alone does not exhaust speech, positioning is also explicitly or implicitly present.
Attention to small forms
First, it is easier for people to generate content in small forms. Secondly, such forms are more likely to compete for the attention of the audience. Third, they are less subject to change - this is important due to the fact that in the process of development of any topic its structure changes. You can, for example, write an article and at the end understand where to begin. When there are ready-made decorated and meaningful “bricks” and blocks of related considerations, it is easier to build them into various larger logical structures.
Discussion + structuring
People are also easier to generate content in the process of communication, discussions, discussions. A good structuring of the topic is almost equivalent to its good comprehension. It is desirable that these two things are harmoniously interconnected. Otherwise, the mountain of produced content will grow dramatically, but it is not enough to add to the global understanding of the topic, to provoke duplication of topics and walking in a circle.
The question is how to connect discussion and structuring at the interface implementation level. For example, the project Rizzoma (the successor of Google Wave) is trying to do this. In this Google Study, I formulated some criticism of Rizoma.
Top media and search: the principle of uncertainty?
The maximum audience size can be achieved in different ways: if we are talking about a news resource, a tape format, then it provides short-term synchronous attention of a large audience; and if the reader comes from a search query, here we have the asynchronous attention of the audience, which at each moment is not large, but the amount for the entire (indefinite) period of time can be large.
The question is whether it is possible and how to combine these two approaches in one resource.
Separation of rankings from services
In the situation of a large amount of content, its ranking is an important factor determining the attention of the audience. If the content can be separated from the services, then the ranking too. For the “Internet content” such a function looks like a natural analogue of search engines ranking websites on the regular Internet. The content ranking service separated from the sites would use not only the information obtained from the site on which the content was produced. In particular, due to the processing of data on the activity of the author on other sites.
Klout has already implemented the idea of ​​separating the ranking of authors from services, and even displays the author's content, but does not make this service for content. And could - ranking authors and content - things are strongly related. Attitudes toward the author depend on what he produces and does, and this attitude is then transferred to his content — from the author with a high reputation, the content initially receives more attention.
Amount
As suggested above, the user of an arbitrary Internet resource selects the text and supplies it with metadata, these metadata serve as the “address” at which the text falls into the necessary thematic section of the “Internet content” and is correctly positioned there. (Actually, the assignment of another separate identifier - the URI - may not be necessary then).
Here it is necessary to fix an important point - usually people do not like to perform extra actions; tagging the text is just an example of such actions, since the user doesn’t really need it, because the target action — publishing content — will happen without it. Getting into the right section, like on Habré, or to your target audience, like in blogs, happens at the expense of other things.
However, if you create a content object outside the context of a specific service, simply as a thing from the Internet of content that can be used in different services, then providing detailed meta-information is the only way to reach your target audience. The author is vitally interested in this.
So far it turns out that “Internet content” is still a specific service, the minimum function of which is to store data and metadata, which it can transmit to other services through the API.
Further, the same or another service ranks authors and content (I have thoughts about the ranking mechanism, if briefly, this is using the PageRank algorithm in the network of heterogeneous objects — content and people in this case; what is considered a voting link is a separate story, for details see the text on the link).
In this embodiment, it turns out something similar to search engines or Wikipedia, because such a resource does not have a specific thematic positioning. But the issuance of a search query - in my opinion it should be completely different. Not only not the same as search engines, but not even like Wikipedia. More precisely, there is a similarity to Wikipedia - probably an overview article for acquaintance with the topic should be present, but you need an even more or less standard description of other important resources - the top authors most competent in this topic, the list of organizations whose profile this query most closely matches, the list sites. And there is no top content, because The review article itself is top content - and here an important difference from Wikipedia is seen - the content gets to the top as a result of ranking over the entire activity of all users on the Internet, and not individual nameless authors of articles for the wiki encyclopedia. I think a ranked list of references is also needed.
In general, what should be the issue - a separate non-trivial question. How different themes can be standardized by one format of issue and so on. For example, the history of the subject is now a typical section of Wikipedia articles (i.e. some standardization already exists there), and the person Pavel Zakharov claims that he knows the only correct way to tell about the evolution of any subject (in the sense of an algorithm to present it somehow). We also recall the above-mentioned problem of combining the news and search tops - probably there should be a news component in the issue, some ranked tape of new publications (divided by positioning).
In general, the “pyramid structure” seems to be correct, where the top of the pyramid is a review article on the topic, and the lower levels (available as an option) cover an ever-increasing amount of various relevant content.
Of course, the idea to create a “better Wikipedia” may seem ridiculous, but think about it - Wikipedia appeared at the dawn of the Web 2.0 era; In the current era of maturity, it is possible to create a more modern product, taking into account the accumulated experience from related areas and the current trends of the future. Nowadays, only selected altruistic fanatics tend to be the author of Wikipedia, but in its modernized version, everyone would rush there - for the sake of a large amount of relevant audience.
It remains to understand how the concept of combining in a single structuring and discussion space fits into this scheme. Is it worth stimulating discussion right in the “new wikipedia”? And where in this picture of the world is a place for other content services on the Internet. The fundamental problem, which was mentioned at the very beginning, is only on the one hand the problem, and on the other - the incentive to develop and improve content services.
It is only clear that the “New Wikis” themselves may at first be somewhat competing, although in the end one will become strongly dominant - the one that will offer the best ranking and issuing format. As now Google dominates in the field of search, and Facebook in the field of social networks (the current Wikipedia does not count - now it just complements the search). Probably, in the case of the dominance of one of the new wiki, other resources will begin to adapt and aim at integration with it (ie, they will begin to resemble the current applications to social networks). The one who is better able to realize combining structuring with discussions will be able to integrate better and mutually with the giant.
It is also possible that Google or one of the current giants will partially evolve in this direction.
PS I already wrote about almost all of this on Habré, but in different articles and, moreover, now I understood one of my mistakes, which earlier prevented me from more or less consistently presenting the topic.
Source: https://habr.com/ru/post/201742/
All Articles