I think the patterns in Facebook's "clumping" algorithms are interesting. This screenshot, I assume, shows that Facebook believes most people are such horrible writers that it only makes sense to look to the second paragraph or line to deduce a writer's point. 

Add to this to the fact that Facebook has recently begun sampling the second paragraph of a blog post to create a link preview and I think we' see the the beginning of a pattern. 

However, algorithms can only parse so much content as wibbly wobbly gibbldy gook. At some point, human beings are going to look at the Facebook presentation and think, "Hey. This is entirely inaccurate."

What can we learn from this?

Observe that algorithmic composition is a process inherently flawed. To pigeonhole these writers as buddhists is patently inaccurate, but to pigeonhole the sentiment they're expressing as buddhist is even less accurate. 

Humans recognize this immediately.