No real tips here, just something I was thinking about tonight that sort of made me chuckle. Back at the turn of the millennium (isn’t it cool that we can say that?) I was working with a couple of guys on a system to measure and track a web site’s popularity. What we were trying to do was provide clients a service that would give them a score for their site based on two major components–links from other sites and search engine rankings. We were going to get mega-rich doing this of course.
The first side of the equation, links from other sites, worked basically the same way PageRank works now. Every page on the web would be assigned a score, and that score would be determined by the scores of the pages that linked to it. Remember, the web was a lot smaller back then. It goes without saying that we underestimated the enormous size of the web just a few years later, and we hadn’t even started to think about the small complexities that make this problem so difficult or the amount of computing power it would take to tackle it. We also had no idea that guys a lot smarter than us from Stanford were doing this and planned on giving it away as part of a huge advertising company.
That’s right–Google’s an advertising company. They just happen to have a lot of stuff you want to look at and use. And they make sure their ads are involved somehow. Google’s search engine just happens to be one of the tools they offer that you like to use.
Search was the other side of our scoring system. What’s crazy to think about now is the number of search engines we were factoring in. I want to say our first prototype was analyzing results from 35 different search engines. Back then, Yahoo! was the big dog, but several other search engines had decent market share. To top it off, the guys who were heading this project up and funding it were in Europe, so we had to factor in the engines that were being used there. We were going after the long tail before we knew there was a long tail.
We were looking at the first 10 pages of results for each search term on each search engine. And we were planning on each customer having 10 to 20 different search terms each. That’s at least 35,000 results analyzed for each customer if you’re keeping score. And we were going to do this once a day. And if a search engine changed the layout of their results? Well, we’d have to re-write our method of parsing those search results. Like I said, we hadn’t really though out all the details.
Having pulled several all-nighters attacking this problem, I’m utterly amazed that now Google gives away their PageRank score not only to site owners but to anyone who cares to look at it. And after all of that time spent writing software capable breaking down pages and pages of search results from different engines, isn’t it funny that all that really matters now is being on the first page (really, in the first three results) of Google?
If you’ve checked out your anaylitcs, you’ve probably noticed that Google is the only search engine that matters, at least for now.
I’m glad somebody got mega-rich on this stuff. I at least feel better knowing it was a good idea. 
Similar Posts:
- Link Narcissism
- Interesting Case for NOT Having a Sitemap
- SEO – A Five Year Plan
- Finally…Let’s Talk About Money
- Search Engine Optimization Round 1
WANT TO READ MORE?