"Vebalta" opened her face

The company “Vebalta” , recently admitted to the state of two promising students , announces the transfer of its main project - the search engine - to the stage of public testing. The development and testing of the system lasted a little less than a year - from August 2005 to July 2006, and now it is available for use.

Vebalta announces ambitious aspirations to become one of the leading players in the Russian search engine market. Olga Arakelova, head of PR and branding of the company, notes that the short periods in which the system was able to be launched allow us to say that innovative technologies were used in its creation. “We have the know-how, but the market is harsh and dictates its own conditions. Unfortunately, we cannot reveal our secrets right now, ”says Olga.

Speaking of innovations, the developers of “Vebalta” mention a new mechanism for indexing documents and a modernized procedure for selecting and sorting results. The search process takes into account the morphology of the Russian language. The combination of these components improves the accuracy of responses to user requests.
')
The Advanced General Purpose Search Engine, as its creators describe the system, was developed and launched entirely at the expense of the company itself. In addition, it is entirely a Russian project. At the time of the transition to public testing, the system indexed about 6,181 GB of information - this is about 227,089,000 documents. Their volume is planned to increase to 1 billion by the fall, and by the end of the year - to include the European Internet in the database.

Along with the release of “Vebalta”, the contextual advertising service “Optimist” is announced. Its launch is scheduled for the end of July.

Igor Ashmanov, managing partner and CEO of Ashmanov & Partners , commented on some of the nuances of the search engine that came out, noting that he did not directly test it: “To test thoroughly, you need to have 3-5 linguists, 100–200 specially selected queries (frequency and rare) and spend 2-3 weeks. This is described in detail in the method of testing search engines Nikolai Kharin, which is easy to find on the network . "

“By the way, we note that“ Vebalta ”for some reason only shows this technique at the end of the first page ,” continues Igor. - There are techniques for “blind” testing of search engine quality in ROMIP, but “Vebalta” has not participated in ROMIP tests so far. Therefore, I cannot give a qualified opinion on the relevance of the search for “Webalts”. I can only give an opinion, a feeling. ”

Feels like Igor Ashmanov, the relevance of the results from the search engine is quite up to par, but the quality of the show suffers. “For example, the method of Kharin from the example above was, in principle, found in the first links, but in the annotations the wrong parts of the text are shown, which creates a feeling of low relevance. But this thing is gainable. ”

“As for the volume of indexation, it is still lower than in Rambler or Yandex by a factor of three or four,” says Igor. - But not the fact that it affects something in terms of the quality of the search, then the question is in the method of selecting sites. In addition, there is no doubt that the big base "Vebalta" one way or another will collect, they have the money, technology, too. "

“It’s surprising yet something else -“ Vebalta ”does not show anything new in the sense of the user interface, streamlining the search results, navigating through them. There is a sense of lack of debut ideas. I talked with Alexei Gureshov, the project manager, so I know that they have fresh technical ideas. But they are not visible yet. ”

To the question of “Habrakhabr”, “is there a place in Runet for a search startup, or is the niche already fully occupied by the main players” Igor Ashmanov replied that “there are plenty of places”. “I myself am engaged in search startups (see “ Applicate ” ), but I wouldn’t do it if I thought that there was no place. The question is what azimuth to take, directions — all 360 degrees. You can, in principle, run along the old road, like a gopher in front of a car in the headlights. Imagine driving three KamAZ trucks - Google , Yandex and Rambler , and you run in front of them. After all, crush sooner or later. Well, for them to run - also makes no big sense. Therefore, it is better to choose your path and gradually turn it into a torus road, on which everyone will begin to ride. ”

“While“ Vebalta ”, in my opinion, moves exactly along the same road as the“ Ramblandx ”. It looks that way. The same endless tape of search results, the same annotations, the same headings. There are no differences visually. Besides the fact that the “guglemblers” still have a lot of all sorts of other search and other services, while the “Vebalt” still does not have them. Personally, it seems to me that the competitive advantages of the “and we are but relevant” type will no longer have a significant effect on the distribution of shares of the search market. We need to show some completely new "chips" and create a convincing legend around them, as Google did.

Igor Ashmanov has not yet noticed the effect of the document indexing mechanism: “As I said, testing is needed, this is time, and the second is subtle differences, to establish which need to be carefully tested, it may not concern the user. Here, I really can be wrong. For example, if “Vebalta” could loudly declare and then convincingly show that there are no doorways and other garbage in it, it would be a strong move. ”