Classified sites should be wary of cool news

The media has recently reported on the Cool News Network (www.kooxoo.com), which has two basic topics: one is to do classified information "vertical search", and the second is the founder of Chen in search of "senior" background. All in all, these reports try to wrap the cool network into a "search concept" website.

Coincidentally, before these reports came out, I began to focus on and track the trend of cool news network. My point of view is exactly the opposite of what the media reports. I think that cool news in the "search" aspect is nothing special. The nature of cool news is not search, but the portal site of classified information (portals); This should be worth the vigilance of various classified websites in China.

First, search?

Let's say search first. Google, Baidu represents a model, it can be said to be a whole network of basic search mode, while Qihoo and Daqi represent another whole network mode. Here I would like to say in detail: from the surface, Qihoo and Daqi are currently the BBS content aggregation, seems to be "vertical search (Aggregation)", but in fact they are based on the wide area of the network search and aggregation-that is, basically for all Chinese BBS; so I don't think of them as "vertical". And the third type of search mode, is the cool network for the representative, the real "vertical search"-they do not index a large number of sites, direct index related information on the host site can be. And according to the business scope of cool, I analysis, such as renting this piece, they index the main Bailai rental website can.

and a real search site, the main needs of three functions: 1, the crawler system analysis of a large amount of unknown Web site Web page, 2, index crawled back through the Crawler Web page, 3, the user search request analysis and response. In the last paragraph, I tried to clarify the current three different types of "search" sites, because these three different search patterns, in the search of three features, the technical content is completely different. The technical content of Google,baidu is the highest, and the technology content of cool mode is the lowest.

Because the first, cool news does not need to use the crawler analysis of unknown sites, only the limited information on the classification of the host Web site analysis, and then crawler-oriented crawl can be; (Cool news whether using machine analysis said, I just explained that the artificial analysis at this point can achieve the same effect); Indexing these hosting sites does not require a special strategy, because several classifications are identified, information is easy to format, for example, rental is no more than the location, rent, contact information and so on; the most interesting is the third-actually the real search site, the user search request semantic understanding, the technical content is very high; but cool news took two " Tricky "way to reduce the technical difficulty in this area: (1)" Limit "The user's query request, in order to reduce the system to the user semantic interpretation of the difficulty. Take the rental as an example, prompts the user to enter the noun "place name", so the system to the user semantics explanation is very easy--but even so, cool news to do is also bad, for example, search "North three ring" has 1033 information, and search synonymous "North three ring" only 143 information; "North three Ring Road" only 7 messages (2) The Web site lists the main query conditions, similar classifications, used to reduce the likelihood of user personalized query requests, thus reducing the difficulty of semantic interpretation of the system. Similarly to rent as an example, cool news through several of the most commonly used classification recommended, in fact, users may not have to search again, only click on these several "classification" can be, the system does not have to understand the user's semantics

Through the analysis of the above, theoretically, in fact, do not need search technology, cool news site can be made out. So cool news is a search concept of the site, looks a little reluctant, more appropriate positioning should be a classification site "mashup lightweight applications."

It is worth mentioning that cool news more proud of some of the "technology", in fact, is precisely the application of mashups. For example, "Pull-box search", that is, the classification and mapbar of the map services mashup; For example, "the world's first" bubble "search engine", in fact, is the crawler crawling hard point, plus the page Ajax can be solved-these "world first" technology, looks "cool", in fact, as mentioned above , really from the technical view is also a lightweight "flower work."


second, the fragile industrial chain

In other words, cool News is the classification of information users portal site, that is, if users want to find relevant information, first to cool, this is the real positioning of the current cool news. This vertical portal positioning, in fact, and Sina's news portal positioning is the same, but Sina in the industry chain "host" is a traditional newspaper, and cool news in the industrial chain in the host is: Classified Web site.
So cool news in the classified information industry chain, such positioning is appropriate?

To answer this question, let's take a look at the two cases that Craigslist, the first website of foreign classification information, encountered: When Craigslist became the largest classified information site (please be sure to pay attention to the background of the industry boss), craigslist allowed housingmaps.com to cite their data, but rejected another oodle.

To this, very coincidentally, I write this blog, search information when found, cool news site side has an article to talk about this problem, excerpt as follows:


。。。。。。 Always have friends care about, cool news This kind of classification information search and classified information publishing site is inevitable antagonism and competition, classification information publishing site will be because of conflict of interest and refused to classify information search site access to their website data?

In a rough look, these two types of sites seem to have a certain competitive relationship, and friends generally cite the U.S. vertical search site oodle by craigslist warning "hope no longer use" information from its list of examples to prove that this statement is very good.

However, after careful analysis, you will find that this is a pseudo proposition. First, the story of Oodle and Craigslist. Why does Craigslist not block other search engines that consolidate its information and only warn Oodle? Just look at the Oodle Web site--it brings together information from multiple categorized lists under one page and looks like the second Craigslist. This is like Google if you crawl Yahoo News, and the form and Yahoo News channel is no different, will be warned by Yahoo. This can only be explained that if the search engine business model and the original information provider too much overlap, rather than an alternative or technical innovation, it is bound to conflict with the original information provider.

Craigslist does not reject the integration of other vertical search engines, such as Housingmaps, and it can also be seen that Craigslist, while enjoying the status of "one brother" in the American classified information publishing community, cannot resist the trend of Internet integration and sharing. From the user behavior analysis of cool News also found that the user is not only in the cool news site, almost every user interested in information, will be clicked to see the original data page. Cool News as a classified information search portal to many classified information publishing community to bring active users.

From the above analysis, we can find that: classified information search and classified information dissemination of the two business models are like a pair of parasitic animals, together with the birth ...

This article can be used as a cool news to explain the location of their own industrial chain of the official article. They also contrasted the difference between Craigslist's treatment of oodle and Housingmap, and considered it a "pseudo problem". But interestingly, this problem seems to be a "pseudo problem", and I stand in the third party perspective, but it is the most important question. Because the cool news to this "pseudo problem" explanation does not convince people.