Google: From Project to PublicGoogle: From Project to PublicOVERVIEWThe rapid growth of the internet worldwide in the 1990’s sparked a technological revolution that continues to shape the world we live in today. This boom brought with it the perception of limitless opportunities and success in the “dot com” world. As a result, entrepreneurs of all kinds took to the internet with their ideas. Some ideas immediately received favorable attention and led to continued investment in those areas. After the initial rush into this new-found gold min, the advantages of the World Wide Web were apparent to all who came to know and love it. While the apparent endless success of opportunities appeared to come to somewhat of a screeching halt, several entities still continued to make the best of the situation. Today, names such as Ebay and Amazon are commonplace in almost every household with a computer and internet connection. But, perhaps even more surprisingly, the name Google has become more than just a silly name with a meaning most people may not know. It represents a story of unbelievable success in a market that did not take kindly to small competitors. Google Inc. is now a major public corporation in the United States, but going back to its inception, growth, and success, we witness a truly compelling story.
The inherent nature of the internet allowed for a vast number of potential successes in entrepreneurship. This relatively new marketplace connected people from anywhere on the globe to virtually any and everything. Information could be easily accessed, obtained, downloaded, or shared at any time of day with just about anyone else who had the means to surf the internet. While many companies sought to reap the benefits of easy accessibility by offering various products for sale, others chose to pursue other avenues to success. Selling products on the internet appeared to be a clear and obvious choice, within reason. Specialty shops, existing brand name department stores, and electronics stores took to the internet to provide consumers an alternative to face-to-face transactions. Still, other entrepreneur’s saw the value of advertising on the internet, and quickly jumped on the opportunity to host and link sponsor web sites at a huge profit. Without heavy regulation, however, the internet saw potential in a completely opposite direction with genres like gambling and pornography. But with all of these potential money making schemes, a much less obvious tool came into play as one of the most successful and stable components of the internet. Google entered the World Wide Web as an internet search engine and surpassed expectations to become the world’s premier search engine and so much more. To this day, Google continues to grow and enter new markets to ensure its viability for decades to come.
During my initial research on companies that sparked my interest, I spent a great deal of time familiarizing myself with not only the financial prowess, but also the stories of how those companies came to be. Perhaps the most intriguing aspect of Google Inc. is the story of its inception. I examined several other aspects of each company, and viewed their mission statements to gain an understanding of what their operations seek to achieve. While most of the companies or entrepreneurs I researched seemed to have goals related and measured by financial success, I was drawn closer to Google after reading its mission statement. According to Google Inc.’s Company Overview, “Googles mission is to organize the worlds information and make it universally accessible and useful.” Not only is that one of the most concisely written and straight-to-the-point mission statements I have ever seen, it is also the most utilitarian, with a goal of enhancing the world today. I was extremely impressed with the underlying theme behind Google’s work, and my curiosity continued to peak.
GOOGLE’S BEGINNINGThe story of how Google started up is unique in the fact that it began as a project. Two graduate students, Larry Page and Sergey Brin came together to work on a research project at Stanford University’s computer science department. At the time they began working together in 1995, they looked into developing a new search technology that would operate more efficiently and on a completely different principle than existing web search engines. At the time, the most common method utilized by the major search engines on the internet was returns based on how often keywords appear in a particular website. The theory behind Google’s search technology approached the subject from a different angle. Page and Brin hoped to produce more relevant results, rather than something based on frequency. Not only was the engine to be highly reliable, it was to be designed to produce results with unprecedented efficiency. Complex results could be posted
The researchers from Stanford followed up with the search engine in order to find and identify a particular page. The way they found this webpage was by applying their unique algorithms. Google’s new algorithm was based on algorithms written in Python. But unlike other search engines, like Google+ or Twitter, it would take far more computing power than previous search engines. And while they could quickly identify the keywords in the webpage, Google began to have problems because they found a lot of errors. The group at Stanford discovered that, at times, the algorithm was completely off by more than 100%. That’s about 25%.
In a paper published in 2010, the group’s co-author, Steve F. Shiller, a professor of computer science at Stanford and an expert on Google’s search technology, wrote, “We have found that more than 90 percent of online searches do not actually return results. This type of error is caused by an enormous amount of information and information that is actually in the database at the time. It is very common, when new information is found, that the algorithm fails, but that doesn’t mean that all online searches were bad, but that it isn’t always as bad. That is why many search results have more error than acceptable, and in fact many errors occur at all times.”
When they started looking at page results during the first six months after the discovery, they came across a lot of misleading and misleading statistics. For example, just 7% of the total searches in the search results page looked correctly. A lot of these sites did not include comments describing how they read web pages that they were searching for. On average, the results at one of the sites did not show the full text of the web page in “Help” form, which is the most common form given to search results. That led the researchers to believe that the search engines were likely to have a problem solving problem—a problem that is very likely to happen in other industries, too.
This kind of article is what is referred to as “reinforcement learning”—in other words, that there is an inherent advantage to learning from multiple things. When a specific type of behavior is learned, that is what separates it from other behaviors.
The team focused on several problems that emerged from this study. Specifically, they found that the best known problem involved using a very low set of inputs that would allow them to easily perform an optimal search. This approach led them to believe that if the training method was efficient across all types of searches, then it would produce the most results. They also found that they were probably using a much more powerful solution that eliminated this type of error, that would have reduced the time to perform a search.
What led back to the problem that they found in the search results? They realized that the best solution was a strategy that only used very small inputs. They realized that this solution may just be a better fit in most situations. They thought that using very large inputs instead of the low set would have better results for different types of search engines.
They further tested a different kind of strategy, known as hypertext search. To study this strategy, the students worked together over a three week period with a computer scientists at Stanford University. This approach allowed the students to perform very large amounts of text search and write long formulas, so that they could quickly create the most important, most common searches.
While some of the students focused on the high-level techniques being used to create better search results, others didn’t get as much attention. The high-level methods allowed the researchers to look at the basic concepts that were known to have a problem solving relationship with the high level methods. That also meant that the researchers could not fully understand