Replies

Sure - let’s talk about speed. Years ago, when I would secure a new domain name (like freerepublic.com for example), it could take a day for the domain name to work. Meaning I could type in the domain name and it wouldn’t work. That’s because the technology didn’t exist to update the networks to let the internet fully know that the domain name was ready to use. That is no longer the case. The point of this example is that a domain name can be accessible in minutes vs days or hours anymore.

The point is that things on the web happen much faster today than they did a few years ago. The different computers that make all of this work are collaborating at amazing speeds for a lot of different reasons.

The next issue is about ‘spiders’ or ‘robots’. Those are little programs (not so little anymore) that are constantly searching the internet for new content. There are two kinds of new content: new and updated. They do their best to identify them. Google uses it to keep its search engine fresh. And they dedicate massive computer networks to the process... and it happens FAST.

So that means that when a user posts something on Reddit or Free Republic... the search engines know about it pretty quickly. The timing can vary based upon the reputation of the page (does Google value the page for example)... but generally those times for even the less valuable content are getting shorter as the tech gets smarter and the systems get more powerful.

Now, when content is added or updated (like this post I’m typing) a spider will notice it and take ALL of the content that is available to it and ‘index it’... frankly it’s a fancy word for ‘copy’ ... kind of complicated really... but we’ll just say they make a copy and now it’s in their database.

Now the speed of the internet and the power of the massive server farms create this ‘snapshot’ of the content on the internet... it’s mind-boggling to think of the scale... and now it sits there to be used by entire other technologies that are designed and created to analyze that content.

Every tweet, every post, every facebook message... all of it’s available to be sifted through by very intelligent and fast programs that can spit out answer. What answer? It depends on what you ‘ask’ it. But essentially what’s happened in this article is that someone has written a program to scour the available data sources that have captured where, when and who wrote it (as it relates to the site it was written on) and then begins to parse it into whatever other information they have about the who and the where and then when to generate that bubble graph.

The tech has gotten that fast and that big.

Smarter people than me will cringe at what I’ve written... but that’s ok. :)