Half-way "In Conclusion"…

As you see, although all existing today search technologies and systems work (and work quite well in places), they still don’t solve the main problem of search to the full extent. When the speed is satisfactory, the quality of search lacks; while  when the search is precise and adequate, it takes lots of time and computing resources. This problem can of course be solved though increasing the level of computing capacity, but stuffing your office with a few dozens of ultra-fast computers, which will continually process phrase queries made of thousands of words, working with gigabytes of incoming correspondence, technical literature, reporting documents and other information – is more than just irrational and inconvenient. There’s a better solution.