How to fill 90% of the free memory? was shared through social media (at least Hacker news and the Stack Exchange twitter feed), and it rose to the top of the Stack Exchange hot question list (which picks on popular questions and makes them more popular). It's simple enough that many people understand what it's about, don't get bored halfway through it, and have an opinion on the answers. That's a recipe for popularity. Quality doesn't really enter into it. This question has had its 15 minutes of fame, and if it wasn't for people specifically browsing the top popular or high-scoring questions, it would fall back onto obscurity.
Convince apt-get *not* to use IPv6 method didn't get very popular overnight. It got onto the Stack Exchange twitter feed, but that didn't amount to much. You can see in the timeline that the votes have been trickling in in the two years the question has been online. It has been linked here and there, and it solves a real, immediate problem that many people face. This is the kind of question that we like to see become popular.
Our top viewed questions are (in close order):
Oddly enough, they don't have a huge score, unlike the next runner-up Why was '~' chosen to represent the home directory?. This may have something to do with the manner in which they became popular: the top three gathered their votes over time, so they presumably got views because they corresponded to problems faced by people searching the Internet, whereas #4 got its views through two discrete events: on the first day, and on the day several months later when it was shared on Hacker News.
It seems that HN readers have Stack Exchange accounts (so they view and vote), but people with actual problems to solve don't (so they view and don't vote).
How to fill 90% of the free memory?
when I saw the scores 12 hours after it was written. There's an answer in there that got 75 when normally it'd be lucky to end up with 2. Really, honestly. Pretty zany.