Faster access to pages… for Google
DNS servers, for people who don’t know the technical details, are basically the phone books of the internet. On the internet, every computer server is known by a set of digits known as its IP address. For example, when you typed tnl.net, the DNS server looked up the name and discovered that it was at 206.127.35.2 allowing your computer to connect to mine.
I’ve argued, for many years, that Google wanted to find a way to access new web pages at a much faster rate. The challenge the company has had is that it is difficult to find new pages when they appear. While traditional technology to discover what pages are available on the internet has evolved and Google has managed to coerce some site owners in providing it with a quick update when changes are available on their site, the efforts the company has pushed have also been helpful to its competitors, who could build on top of the processes and open standards Google was fostering.
With Microsoft getting some early successes in the search game with their new Bing offering, and new entrants in the real-time search business eating up some of Google’s mindshare in search, the company needs to do something radical, lest the cornerstone of its business, and the source of most of its revenue, be undermined.
Enters the DNS system. Every time a page is called, your browser makes a DNS call (several, in fact, as every web asset can result in a different one). In other words, the DNS system truly serves as the heartbeat of the internet and convincing a large swath of users could allow Google to get an idea as to what’s new on the internet.
If, for example, a user were to access a new page that’s not in the Google index, Google’s own DNS servers could be wired up to alert its search spiders to immediately pick up the page, analyse it, index it, and make it available to its search users within seconds of the page first being accessed. This could give Google a substantial advantage over Microsoft and others in indexing the web in real time.
Some potential risks
There is, however, a huge caveat in all this. For starters, Google needs to convince a large number of people to access their DNS. Providing the product for free may work for some but will not convince everyone. Another issue they may have to deal with is the perception that they might snoop on personal data (something that is already being addressed on their site). The ability to access information about everything you do on the internet, whether it is via a web browser or another application like Skype, online games, or video and music player, is granting Google some brand new capabilities and not everyone may be willing to share such information.
Google will also have to contend with large numbers of potential denial of service attacks (or worse), which have become more common of late, against those DNS servers. Such attacks could represent a substantial reputational risk to the company. If, for example, one of Google’s DNS servers could be compromised, the hacker could decide to redirect the traffic of banks or other financial institutions to their own sites. The potential financial impact of such a thing would become a legal nightmare for the company.
The Prize
All this, however, can be counter-balanced by the rich prize the company would get in being able to index every bit of internet content within seconds (or even nanoseconds) of such content being available on the internet. If that were to be achieved, the company’s perch at the top of the search heap would be guaranteed for a long long time and its continued dominance in the advertising world, based on the rich analytical data it could get from snooping on users of its services, would provide it with cashflow that other players on the internet would have a hard time to match.
Furthermore, Google could have control over where people go and could, if they decided to be evil, redirecto such traffic. That would be a tremendous amount of power.
All told, an interesting move and it will be fascinating looking at where they are planning to take this.