An article on doubles mode in DDR might be interesting in HN's sense, or it might not. If it were just a basic tutorial, that probably wouldn't be—tutorials are useful to people who are working on a specific task, but have little appeal to the general reader. If, on the other hand, it went into the design of doubles mode, the history of it, why it is the way it is, what makes it harder than you might expect, clever things that have been discovered about it—all of that would be on topic. Something that could go either way might be a story about, say, a couple who met because they were good at doubles mode in DDR. That would depend on how good an article it was.
This is important because the kind of story that gets excluded this way—i.e. the kind that people find interesting but assume is not "hacker-related" enough—can often be the best kind of HN submission. Why? Because it's unexpected. It's the one you can't predict from any sequence.
(Edit: Unpredictability isn't the only quality that matters, but it's the rarest. Actually, maybe it is the only quality that matters. For a while I was thinking otherwise, because a front page consisting of random numbers would be unpredictable, but not interesting to read. On the other hand, it's unpredictable only along one axis. From another point of view it would be fully predictable: random numbers again?)
Keep in mind that finding something interesting isn't the same as liking it. Also, when we say "interesting" we mean intellectual interest, not all kinds of interest or curiosity. For example, there is social curiosity (the sort that powers celebrity gossip). There is political curiosity (wanting to know how one's side is doing against the other side). There is sexual curiosity (no comment needed). These things all have their place, but not here. On the other hand, there can also be overlap with intellectual curiosity, in which case it's fine, though the bar is higher in some cases than others.