When I do serious editing I am usually working from a book or newspaper article and I have the reference at hand. It would often be very hard for someone else to find that passage just going from whatever I put into a Wikipedia article. So it is easy to put in exact references.
I, by contrast, almost never do such editing, and often work with articles of low quality (eg, in the fields of skiing, snowboarding, rock climbing, computer games...). There is often a lot that can be done to improve the article by restructing, rewriting, removing POV or adding snippets of general information - even with limited or no knowledge of the subject. People like me would probably throw their hands in the air and stop working overnight if required to find references for everything.
The thing is, for many people, editing is *fun*. I actually honestly gain pleasure from taking an unstructured 1000 word article and turning it into a 500 word structured one. Finding references is *work*. If you know something to be true, to find a reputable refernce to back you up is simply hard work in many cases. "no original research" says that if something is true, it should be easy to find a reputable reference - well, it isn't always.
There really should be different sourcing guidelines for different fields in Wikipedia - popular culture is just "different" to history, science or geography.
I also suspect that a guideline could say somewhere that if it's possible to verify something on google, then that may be good enough in some circumstances. As opposed to making a claim that cannot be verified even by someone searching the entire internet.
Steve