Although the post is about the traditional training of a non-traditional topic, your points are well founded.

In the conclusion of Alan Mislove’s Doctoral Thesis “Online Social Networks:
Measurement, Analysis, and Applications to Distributed Information Systems”. Dr Mislove explains

“However, the content shared in emerging systems like online social networks is different from previous systems: the content items rarely have links to other content items; rather, the links connect the users themselves. Thus, a new approach to finding relevant information is needed that can compute the reliability of a given piece of information based on the combined reputation of the users who created or endorsed it.”

That’s huge! What he’s basically saying is that search engines calculate the importance of content by the quality and relevancy of the links that page of content has to other forms of content. But in a social network content increases in quality and relevancy by being connected or endorsed to users that have a popular profile or a lot of connections.

Jeff Weidner
Social Media Swarm
http://socialmediaswarm.com