Eli Pariser, TED 2011
Speaker Profile: http://www.ted.com/speakers/eli_pariser.html
The TED talk highlighted what I’ve noticed everywhere in Google-search, buzz and Facebook wall feed-results only from a few friends I commonly follow. It really hit me when I couldn’t see posts from Zubin in Buzz and when I realized that I have 300+ plus friends in facebook but don’t see updates from majority of them.
And what is said in this talk is completely true-the internet is really not what I wanted it to be. To me, just like the speaker, I wanted the internet to expose me to new ideas that I had never thought were possible but were something that essentially should know. I realized this was why I couldn’t see Zubin’s post after seeing that I couldn’t see Zubin’s posts while a friend could. This was also the reason why I used to see posts from a few select set of friends while I saw nothing from most others and even thought “How can people survive without using facebook?” :D.
The implications of such an algorithm based filtering of the content that will be shown to us is vast, the most important of them being the fact that we will see only what the internet thinks we should see. For instance, I’m a person who likes to know about various humanitarian crisis all over the world. However, I could see no such result in my Google searches or my social networks.
This used to be the case with newspapers and their editors but then the case has slowly improved in their case with the development of journalist ethics and more. But, still the internet has a yet to adopt such an ethic and highlight information that we should know rather than what it thinks we should know. For instance, knowing about the latest video game is very essential but then it’s important to be updated about the latest developments in the world.
As rightly said in the talk, the web should highlight ideas that we like as well as the ones that challenge us, news that interests us as well as challenges our very understanding and interests. A right mix of both, a balance is very much essential else we will end up learning more about something we already know quite a bit about and nothing about everything else.
This is certainly a big issue and there is a dire need to come up with algorithms that can reflect such features else the internet will become the exact problem that it was intended to solve.