“Hmmmm, car window is sticking. I wonder if there’s a way to fix it without taking the door panel off.”
Googles “car window sticking.”
Finds article titled “How to fix a car window without taking the door panel off.”
Clicks on link.
Third step: “Take the door panel off.”
I’m not sure how we expect Google to circumvent human deceit via algorithms. (After all, not only was content creator actually lying with the title of the article and page, it was part of a network of human-designed and machine-assisted sites linking to each other, pushing the page rank up.)
I’ve noticed recently that this is becoming an actual problem for some of the more obscure searches. Searching for an error message while coding results in dozens of sketchy sites copying all content from the one site that mentions the error, burying the original site. Why? Because computing power makes it trivial to set up those dozens of sites that obfuscate the useful one, and the added-up pennies from ad revenue matter when you’re talking about millions of pages. Linking from reputable sources doesn’t help as much if people aren’t linking to the search term in the first place.