Google Roadmap

Written for Spider Magazine October 2005 cover story. Also available through spider’s website.

Google’s Feeling Lucky

In just seven years Google has managed to come in from behind and completely dominate the online search market. Can they now do the same in other areas or will they falter in their quest to organise the world’s information?

Google founders Larry Page and Sergey Brin met in 1995. By 1998 these top two students in their class at Stanford had bought a terabyte of storage at discount prices and built their first data centre in a dorm room. Come December of the same year, they moved operations into a garage and were answering 10,000 search queries per day. There was no turning back after that and by June 2000 they had the largest search index while handling 100 million queries per day. So began Google’s quest to organise the world’s information and their insatiable hunger for data.

In 2001, they appealed to the geek chic by organising over 500 million Usenet archives (the internet’s largest bulletin board/discussion network) making it entirely searchable and manageable. Word, Postscript and PowerPoint documents followed soon after, succeeded by images and scanned, searchable shopping catalogues. By this time, not only had they already surpassed every search engine out there in breadth and depth, but also in the quality of search. All this in a matter of three or four short years.

The 80 billion dollar (Google’s worth) question is how did they do it so well? And why is it that no one caught up? One reason may be their cheap data centre architecture, which has scaled virtually unchecked. These data clusters are the muscle behind Google and amongst the most powerful computing clusters in the world, unparalleled in their ability to devour inconceivable volumes of data in the order of microseconds without a single hiccup. Add to the mix their proprietary, democratic search algorithm (Page Rank) and it appears they can immortalise the search business.

Google has never been one to walk at the pace of others. Historically they have taken a sword through conventional business and marketing hype. They are driven by an academic zeal and childlike curiosity. Their culture promotes employees to spend 20 per cent of their time exploring personal projects and ideas. In 2004, they opened up a development office just seven miles from the Microsoft Headquarters (where about 40,000 of the world’s top software developers reside). The office which accommodates about 200 people managed to drain some of Microsoft’s key talent.

After dominating the search engine market, Google had to find other occupations for its developers to stay busy and so it went. They started off with Google News which scanned all popular news sources out there, consolidated the news based on keywords which allowed it to determine which news items were referring to the same story and compressed everything into one highly usable page. It changed the way a large majority of users get their news. Here, they ran into a problem: any company out there has to make money from its products (explicitly or implicitly by driving users to their other money-making websites). The issue was that Google could not run ads on Google News since the news stories being displayed were the property of the respective news publishers. Google did not seem to care much though and appeared content with the goodwill it generated.

Froogle, Google’s product search engine was launched around the same time allowing users to quickly compare and contrast shopping prices and certainly raised some concerns with companies such as Amazon that primarily dealt with selling products and to a lesser extent with companies such as e-Bay which provide auctioning services. Once again, they were among the first with a mainstream product search engine.

In 2003, Google acquired one of the most popular blogging services: Blogger. Once again, their mission was to get into the information medium and blogs are quite talkative. Around the same time, Google launched an updated version of the Google Toolbar (released originally in 2000) that included an advanced feature which let users set an option to allow Google to see the URLs of every site they visited. In early 2005, the Google Web Accelerator trumped this feature and allowed Google to not only see the URL and content of every site a user visited, but also store that same content on their servers. This included private pages such as e-mail from other sites, shopping sites, web-based chat rooms and so on. This was their worst misstep and within a week, Web Accelerator was taken offline (see “Don’t be evil” in Spider, June 2005).

On April 1, 2004, GMail was announced boasting one gigabyte of storage along with Google’s plans to open an office on the moon. This led most people to believe it was all an April Fool’s joke (the competition was still offering 2MB to 5MB of email storage). The news was real and the world took notice. For Google, it was only data. They archived it, scanned and showed ads which allowed the service to be provided free of cost. With GMail also came the first wave of controversy for Google that scanning users’ e-mails is intrusive and more profound arguments were raised about Google archiving users’ most personal electronic communications over the course of their entire lives (currently GMail is growing continually). The concern is that if a court order is ever issued, Google will be obliged to turn over private user data to the authorities and only then does one begin to see a minefield of privacy issues. And then again, what if Google servers get hacked? It boils down to trusting a corporation to retain some of your most personal documents in their care.

In December 2004, Google, still hungry for information, announced plans to digitally scan the books within the libraries of Harvard, Stanford, University of Michigan, University of Oxford and the New York Public Library, making them searchable for users worldwide. The task was daunting but extremely exciting. Years of legitimate, scholarly works would become searchable by anyone. Once again, the idea proved to be naive (or maybe Google knew and wanted to test the grounds). Scanning copyrighted intellectual property and putting it up freely on the Web for anyone to search through would mean that the authors would not be making any money for their work. Google came under legal fire and the project was placed on hold until the legalities surrounding the project were resolved. Toward this end, Google has issued a notice to all publishers that wish to opt out of the program to do so before November 2005. If by this point Google has not received any word from these companies, it plans to go ahead and scan the books. In the meanwhile, Google is resorting to scanning works on which copyright has expired.

Two months prior, in October 2004 Google acquired Keyhole Corporation, a digital satellite imaging company that would eventually spawn the technology behind Google Maps and Google Earth. Featuring remarkable ease of use, and as of June 2005, covering the whole planet, Google Maps and Google Earth opened up the planet to users in an entirely unprecedented way. Of late, however, an increasing number of defense personnel worldwide have raised concerns about Google making readily available sensitive security data such as detailed satellite images of military bases, high profile landmarks and nuclear power plants. They argue that these images could provide potential terrorists with an easy tool to plan attacks.

Recently, Google launched Google Talk with one-click VOIP services allowing any two parties to talk with a microphone and headset – such a service has long been a part of instant messenger (IM) programs offered by AOL, Yahoo! and MSN. However, Google plans to take this to the mainstream by allowing cheap or free calls to regular phones. How they plan to make money from such a service is anyone’s guess but it should certainly raise eyebrows at telecom organisations the world over considering this threatens their core business.

The roadmap for Google has so far experienced quite a few bumps and detours. The company that once seemed invincible is now proving fallible after all. The road ahead for Google to extend beyond search is much harder. As far as domination of the search engine wars is concerned, Google can easily claim victory. Everything else Google attempts is, however, an aside and in competition with other similar products offered by companies that have been in the business far longer.

Web Accelerator failed to get off the ground. Hotmail and Yahoo! Mail retaliated to GMail which is currently still in beta. Google Maps is challenged by MSN Earth. The IM markets are dominated by AOL, Yahoo! and MSN and there is a slim chance Google will be able to gain a sustainable advantage in these markets like it did in search. They are gasping for a breath of fresh air. Search and information organisation are harder nuts to crack but the competition will catch up. Google realizes this and it remains to be seen if they hold a trump card.