People's search and its prospects
In a review article by Arthur Welf dated 07/07/2008, devoted to search engines, there is also a discussion about “people's search”. Judging by the few materials that search engines give out (including on Habré) for requests like “social search”, “people's search”, “user search”, something radical in this area has not happened since ( udt : Fresh habrastatyu with an overview of this topic). But it is possible to evaluate the development dynamics of the most famous project in runet - Flexum , in which users are used to improve the quality of thematic search: at the time of writing the article by Welf on Flexum.ru, 6190 authors were registered, who created 3729 thematic search engines. Now there are 8096 and 5707, respectively. There is also a note on Seonews of May 24, 2007, which says about the beginning of the work of Flexum, at that time there were already more than 2000 authors and more than 600 searches. Total for the first about 13 months, the number of users increased by about 4,200, searches for 3,100, and over the next 8 months, the increase was 2,000 and 2,000. Both absolute numbers and growth cannot be called impressive, and the growth rates, if they change, are insignificant. Curious is the motivation that leads, according to Ashmanov, to use Flexum: “If a person is an expert in a topic, then he, as a rule, seeks to confirm his expert status. Our users make thematic search engines in the areas in which they are experts, not for themselves - they already know very well where to look for information - but for other users, and the fact that a person has created a thematic search engine that searches for all relevant information sources. on this topic on the Internet, confirms the expert status of its creator. "
As a rule, a person seeks to obtain or confirm expert status not among all people in general, but only among a community of experts who alone can fully appreciate in a specific subject area. Therefore, yes - if someone creates a search engine, then most likely not for himself or for other specialists, but for some other incentive. For example, it should be a competitive niche in the market for massively sought-after information, such as car searches. Or an area not covered by official science, such as esoteric. But experts from a huge variety of subject areas will not be sufficiently motivated . The project is helped by the fact that authors can embed a search engine on their sites, i.e. in the right environment for them. But not all users of Runet have such sites, and even more, not all experts in various fields are advanced Internet users. I would even say that the majority is not; The main medium of their self-realization is traditional society, they do not blog, do not participate in forums, etc., and use the Internet in the simplest formats, such as e-mail. In this regard, the policy is logical to provide authors with more opportunities within the Flexum service itself, for example, it is possible to create and customize your own online portals. But it is also focused primarily on advanced Internet users.
Is it necessary to conclude that the popular search is not very promising? Perhaps, Ashmanov and K have enough resources to develop this direction, even despite the difficulties, and sometime in the future to succeed. Unless, of course, their cake in the Russian market has been eaten up by Google before, after the model of the service Google Custom Search has made Flexum. It is interesting, by the way, to analyze the development of this Google service, but there are no statistics at hand and it is more difficult for me to analyze the situation in the West. However, I think that the main considerations are the same. The only difference is in the financial and other capabilities of Google, as well as in the size of its user audience - taking into account these factors, it can afford to experiment and grope further steps in the darkness of untrodden roads. Then, should it be concluded that this is not a case for “garage” startups and only big players can do it? And here I return to my sheep - to the “objects + communication” model and factors of social realization of users .
')
I want to show that, at least in terms of creating a content and metacontent basis for the work of search engines, such startups can be successful. I will highlight two basic, in my opinion, conditions. 1) The service should be massive, and for this the activity of the users should be expressed in their usual formats, they should have their usual goals and motivation, first of all, self-presentation and communication. In this regard, for example, creating your own search engine or portal is an unusual thing, unusual and is outside the context of the activity of most users. However, the mass blogging was also not so long ago unusual and unusual, but it either boiled down to familiar things - communication, or actualized the hidden and not found sufficient way out need to declare oneself in every possible way that the web diary allows. On the other hand, the need to systematize the material, the transition from a certain number of copyright considerations in discussions to articles and books is natural, it is present in “ordinary life”, but does not have, in my opinion, a suitable format on the Internet. The object + communication model provides this format — users create content objects that, for example, can be linked by a consistent causal link of replicas in a conversation (i.e., it is easy to implement normal communication), but each of these objects can also be to link various types of connections with other objects, classification hierarchies, sets, topics, people, organizations, and so on. 2) The service should provide an opportunity for "serious activity." To such can be attributed the meaningful aspect of various professional fields. The experts about whom Ashmanov speaks are in these areas and they should see in the service a convenient working tool for discussion, reflection and systematization. The usual social Internet network is not sufficient for this - to turn into a working tool you need to combine various resources into one network (read content objects) and the ability to structure (link) them according to different criteria and characteristics. Which, by the way, is also usually for a traditional society, where a person is habitually in the context of many relationships at the same time, and, for example, the format of various organizations performing “serious activities” combines human, informational, financial, technical and other resources at the same time. Within these organizations, it is possible to confirm expert status and career growth. Depending on the definition of the types of objects created, the types of connections between them, the system of user polls and some other things, the “objects + connections” network can be configured differently, in particular, to create an environment that motivates both the average mass user and rare ones for the Internet. real ”experts on creating quality content.
The result will be, in particular, a high-quality content-meta-content base. And here we come to another consideration. I do not have statistics, but personally from my own experience I will assume that a significant part of requests to search engines on any topic are “fact-finding”, i.e. In fact, they express a desire to get the best survey material on this topic. And this is actually the format of an encyclopedic article (I think this is one of the reasons for the success of Wikipedia - it fell into the sought-after format). What is the best material? - this means that he is actually singular, i.e. This is the embodiment of the ideal of any search engine. As a result, users do not so much increase the efficiency of the search engine by specifying links between content units, as they fall into the sought-after search format by creating the best subject materials. What is important for the "garage" - the created environment does not have to be initially large. At the beginning, it can even be considered just as a convenient individual means of organizing materials, which is done in the non-network project TheBrain , then as a convenient environment for the existence of a small community, such as forum ones. Then several thematic communities and so on, until it grows into a competitor of Wikipedia, LJ, Proza.ru and Habra at once :)
In conclusion, I would like to note that if you look at the comparatively far perspective of the evolution of search engines, you can see their transformation in the future into a kind of certain types of modern expert systems: specifically ordered content (knowledge base based on the “objects + links” model) + logical inference machine = expert system (artificial intelligence). This is the main potential of the “objects + communications” network from the point of view of the actual machine technology - they provide the logical inference engine with the necessary base. In this sense, Ashmanov’s interest in the Internet implementation of expert systems is logical, which is quite possible at the level of individual narrow thematic areas now. But now such bases are created by specialists manually, whereas my idea lies in indirectly attracting user-generated knowledge base to this web-interaction activity.
As a rule, a person seeks to obtain or confirm expert status not among all people in general, but only among a community of experts who alone can fully appreciate in a specific subject area. Therefore, yes - if someone creates a search engine, then most likely not for himself or for other specialists, but for some other incentive. For example, it should be a competitive niche in the market for massively sought-after information, such as car searches. Or an area not covered by official science, such as esoteric. But experts from a huge variety of subject areas will not be sufficiently motivated . The project is helped by the fact that authors can embed a search engine on their sites, i.e. in the right environment for them. But not all users of Runet have such sites, and even more, not all experts in various fields are advanced Internet users. I would even say that the majority is not; The main medium of their self-realization is traditional society, they do not blog, do not participate in forums, etc., and use the Internet in the simplest formats, such as e-mail. In this regard, the policy is logical to provide authors with more opportunities within the Flexum service itself, for example, it is possible to create and customize your own online portals. But it is also focused primarily on advanced Internet users.
Is it necessary to conclude that the popular search is not very promising? Perhaps, Ashmanov and K have enough resources to develop this direction, even despite the difficulties, and sometime in the future to succeed. Unless, of course, their cake in the Russian market has been eaten up by Google before, after the model of the service Google Custom Search has made Flexum. It is interesting, by the way, to analyze the development of this Google service, but there are no statistics at hand and it is more difficult for me to analyze the situation in the West. However, I think that the main considerations are the same. The only difference is in the financial and other capabilities of Google, as well as in the size of its user audience - taking into account these factors, it can afford to experiment and grope further steps in the darkness of untrodden roads. Then, should it be concluded that this is not a case for “garage” startups and only big players can do it? And here I return to my sheep - to the “objects + communication” model and factors of social realization of users .
')
I want to show that, at least in terms of creating a content and metacontent basis for the work of search engines, such startups can be successful. I will highlight two basic, in my opinion, conditions. 1) The service should be massive, and for this the activity of the users should be expressed in their usual formats, they should have their usual goals and motivation, first of all, self-presentation and communication. In this regard, for example, creating your own search engine or portal is an unusual thing, unusual and is outside the context of the activity of most users. However, the mass blogging was also not so long ago unusual and unusual, but it either boiled down to familiar things - communication, or actualized the hidden and not found sufficient way out need to declare oneself in every possible way that the web diary allows. On the other hand, the need to systematize the material, the transition from a certain number of copyright considerations in discussions to articles and books is natural, it is present in “ordinary life”, but does not have, in my opinion, a suitable format on the Internet. The object + communication model provides this format — users create content objects that, for example, can be linked by a consistent causal link of replicas in a conversation (i.e., it is easy to implement normal communication), but each of these objects can also be to link various types of connections with other objects, classification hierarchies, sets, topics, people, organizations, and so on. 2) The service should provide an opportunity for "serious activity." To such can be attributed the meaningful aspect of various professional fields. The experts about whom Ashmanov speaks are in these areas and they should see in the service a convenient working tool for discussion, reflection and systematization. The usual social Internet network is not sufficient for this - to turn into a working tool you need to combine various resources into one network (read content objects) and the ability to structure (link) them according to different criteria and characteristics. Which, by the way, is also usually for a traditional society, where a person is habitually in the context of many relationships at the same time, and, for example, the format of various organizations performing “serious activities” combines human, informational, financial, technical and other resources at the same time. Within these organizations, it is possible to confirm expert status and career growth. Depending on the definition of the types of objects created, the types of connections between them, the system of user polls and some other things, the “objects + connections” network can be configured differently, in particular, to create an environment that motivates both the average mass user and rare ones for the Internet. real ”experts on creating quality content.
The result will be, in particular, a high-quality content-meta-content base. And here we come to another consideration. I do not have statistics, but personally from my own experience I will assume that a significant part of requests to search engines on any topic are “fact-finding”, i.e. In fact, they express a desire to get the best survey material on this topic. And this is actually the format of an encyclopedic article (I think this is one of the reasons for the success of Wikipedia - it fell into the sought-after format). What is the best material? - this means that he is actually singular, i.e. This is the embodiment of the ideal of any search engine. As a result, users do not so much increase the efficiency of the search engine by specifying links between content units, as they fall into the sought-after search format by creating the best subject materials. What is important for the "garage" - the created environment does not have to be initially large. At the beginning, it can even be considered just as a convenient individual means of organizing materials, which is done in the non-network project TheBrain , then as a convenient environment for the existence of a small community, such as forum ones. Then several thematic communities and so on, until it grows into a competitor of Wikipedia, LJ, Proza.ru and Habra at once :)
In conclusion, I would like to note that if you look at the comparatively far perspective of the evolution of search engines, you can see their transformation in the future into a kind of certain types of modern expert systems: specifically ordered content (knowledge base based on the “objects + links” model) + logical inference machine = expert system (artificial intelligence). This is the main potential of the “objects + communications” network from the point of view of the actual machine technology - they provide the logical inference engine with the necessary base. In this sense, Ashmanov’s interest in the Internet implementation of expert systems is logical, which is quite possible at the level of individual narrow thematic areas now. But now such bases are created by specialists manually, whereas my idea lies in indirectly attracting user-generated knowledge base to this web-interaction activity.
Source: https://habr.com/ru/post/56111/
All Articles