I don't think "defining" captures what we can do. "Meandering around" is more like it. But here are a couple observations.
A semi-objective thing about most of the uninteresting, non-HN-appropriate stories is how predictable they are, i.e. how repetitive of previous instances of the type. The more predictable a story is, the less information it adds and the less interesting it is.
There's a reason we say "intellectually interesting" to describe the stories HN wants. There are different kinds of interesting. For example, we all have an interest in the things we're personally invested in or identify with, but that is not intellectual interest.
Similarly, as humans we're highly interested in what other humans are up to, but that kind of social curiosity—think gossip and celebrity news—is not intellectual interest either. So a story can be interesting without being a good fit for HN, quite apart from not all having intellectual interests in common.
Maybe this explains why upvotes alone don't work for a site like HN. We upvote the things that excite our interest. It's too much work to pause and ask ourselves what kind of interest we're experiencing. And the non-intellectual kinds have an advantage—they require less attention to get excited. (For one thing, they rely on recognition, while intellectual interest relies on curiosity, which is an absence of recognition.) This is compounded on the internet, where attention is fleeting and fragmentary.
Oh absolutely, this is a really neat conversation that I was hoping could happen. You mention how predictable stories are, and we all have those moments on social media sites where we're all rolling our eyes. I think it might be interesting to look at HN from something like Google Trends, wherein tracking the topics that are upvoted/downvoted. Maybe do some penalties for topics that have a large front-page success rate and upvote/downvote ratio (and throw in comment # too, same with comment length) and you could start to probably see trends for circlejerk material. We'll always have people flagging stories for duplicates or the like, but certain topics always get through by virtue of many people submitting about that topic.
The advantage to so something like Reddit is if a topic is talked about too much or not at all, you can just add a channel and grassroots it. On HN, you only have a single timeline and front-page (yes, there's searching, but people are lazy). News sites have channels but editors which controls the stream (and most restrict who can edit). I think a more algorithmic approach to provisioning up and down channels for specific topics could be neat. Not really sure where I'm ultimately going with this, kind of free-forming based off issues I've seen with internet discussions and IRL discussions and recent controversies as of late. I think it's important to look at how HN does it's news/info vs. reddit vs. twitter etc, to me it doesn't feel like we've really 'optimized' it yet for user distributed news ... or maybe I'm not hip with the latest app.
A semi-objective thing about most of the uninteresting, non-HN-appropriate stories is how predictable they are, i.e. how repetitive of previous instances of the type. The more predictable a story is, the less information it adds and the less interesting it is.
There's a reason we say "intellectually interesting" to describe the stories HN wants. There are different kinds of interesting. For example, we all have an interest in the things we're personally invested in or identify with, but that is not intellectual interest.
Similarly, as humans we're highly interested in what other humans are up to, but that kind of social curiosity—think gossip and celebrity news—is not intellectual interest either. So a story can be interesting without being a good fit for HN, quite apart from not all having intellectual interests in common.
Maybe this explains why upvotes alone don't work for a site like HN. We upvote the things that excite our interest. It's too much work to pause and ask ourselves what kind of interest we're experiencing. And the non-intellectual kinds have an advantage—they require less attention to get excited. (For one thing, they rely on recognition, while intellectual interest relies on curiosity, which is an absence of recognition.) This is compounded on the internet, where attention is fleeting and fragmentary.