Ask HN: What are your heuristics for deciding which HN links to click on?

i think i used to have some inherent heuristics around avoiding content that i assumed would be a little too technical etc, but at some point i kinda dropped all preconceptions and decided to click on random stuff and exit within ~30 seconds if it seemed boring -- i think it's exposed me to a lot of content that has actually been super interesting that i'd otherwise have avoided for fear of not understanding much.

- subjects that I know nothing about

- subjects I work on daily

That covers everything so needless to say it can be quite addictive.

I click on most things that start with "10 ways to..." and then immediately regret it. I then repeat the process with "Kubernuts (is/isn't) the best thing ever" and then immediately regret it. And then I click on the comments of any article that I have industry specific knowledge on and tell everyone why their opinion is wrong without reading the article.

Gosh I need a new way to kill time.

If its anything I expect their will be fights and complains. Hell yeah inam clicking those l. For now my mind looks for

1. Uber screwed up in a new way.

2. How interview process sucks.

3. Why this X language / framework sucks or we are leaving this.

Hacker News is basically slowly but surely converting me more into a technical person. When I came here first, I mostly looked at the psychological articles and some web articles. But when I felt a bit more focused and bold I'd go for more technical articles. I have been doing this ever since and I have noticed there are two types of technical articles I really like:

1. any computer graphics reverse engineer article, or reverse engineering regarding a game console, or the posts about Dolphin the gamecube emulator.

2. any security article. Especially about exploits or vulnerabilities.

I was surprised at how well I understood them. But I think in part it has to do with that I sometimes venture out and expand my comfort zone. I keep on expanding my comfort zone through Hacker News, it really supplemented my computer science education quite well.

I also use it a lot as a search engine to find high quality educational content. All those upvotes matter. And if the upvotes don't matter -- the educational resource is quite bad -- then there's always some insightful comment about a good resource. I learned a thing or two about deep learning this way and how it relates to topology. I've never read anything about topology before! That's just really cool that some resources can give you a basic intuition about it.

- Golang.

- “X is dead/dying” just because I want to see if there’s any sort of logic to it and to see the comments (which can actually be pretty useful)

- Psychology-related things (especially about depression, PTSD, burnout)

- Tutorials that I may find useful

- Security/exploits (like the latest ESLint issue)

- Apple-related articles (because I use Apple products but know they aren’t perfect and want to see both sides)

I look for keywords regarding the tools I use and that describe myself. For me, it's C, C++, random Javascript libraries, databases, certain computer hardware, companies I use, and certain startup business terms. It can be a keyword soup that doesn't make sense and I'll still click on it.

I also click on words that appear to be made up, because I've been trained that I'll usually get to read about a cool new library, programming language, or programming tool.

In your example, "quantum" is too vague. Even though I've worked for years in quantum-whatever areas, I feel that the word "quantum" doesn't describe myself due to being too general, so it doesn't trigger my attention.

So basically, I just scan the front page for words that are attractive to me, never fully reading headlines until I've already decided I'll click on it.

- Technical things that have nothing to do with how amazing or bad Rust, Golang, or a new js fotm is. Hype distortion is bad.

- Math stuff

- Bug fixes and security things

- Career related things

I try to avoid flamewar topics and articles that debate gender/race/politics/psychology for the 309838th time. They usually just make me angry, everyone has the same opinion coming out of it as they did going in and when I'm done reading I've learned nothing new. I wish they would be posted less.

Usually the comments will help me decide to click or skip.

I don't consciously or systemically apply any heuristics, but if I think back on the topics that tend to pique my interest: Business and VC, economics, infosec, tech politics, FL/OSS drama, psychology / human nature, and oddities that seem out of place here (because those stories must be especially interesting!). Occasionally web dev tools if they look like something I might use — for example, new static site generators.

I nearly always read the comments first and rarely click through to the original story. The comments are the point of HN for me.

1. anything crystal / ruby related

2. rust and golang related stuff, so I can further justify my choice of crystal

3. security exploits

4. dramas, like angry co-founders duking it out on HN, love that stuff

5. philosophy of mind / consciousness related stuff

6. stuff that tells me when I can get a consumer-grade quantum computer (never happens, or can only multiply 3 * 5 for 50 million dollars)

7. anything that sounds like a horizontally and vertically scalable database solution -- been looking for something like citus or google cloud spanner, but with a $5/month starting plan since forever ago. Note: a $90/month starting plan != scalable imo, as it doesn't allow me to justify developing with it.

8. "here is a super fast hash table implementation"

9. "new data structure" (happens almost never)

10. "programming language performance benchmark"

11. microsoft/google/facebook/apple/amazon hating never gets old

12. news that [language I care about] now runs on web assembly (never happens)

13. "RSA encryption broken by efficient factorization of large semiprimes" (world would end for a few days)

14. proof or disproof of p = np

15. autonomous driving stuff

things to avoid:

1. anything related to react/vue, though I might click if it's vue just for them sticking it to react

2. anything about VC stuff, because I don't come here to read that

3. news that [language I don't like] now runs on web assembly (every day)

4. x new javascript framework

Follow your bliss.

I've found that when there are a decent number more comments than upvotes (say 1.25+ comment / upvote ratio) it is generally a controversial topic, probably with a huge amount of subcomments under the top comment. If it is in my domain, I tend to ignore it, as I've often heard both sides many times before and its draining to see debates play out again. If it isn't in my domain and it is interesting, I'll check it out and often learn something. If its outside my domain and not a particularly interesting topic, if i'm bored i may check it out to entertain myself for a bit

for Ask HN, the comment / upvote ratio seems like it indicates a topic that people are interested in and want to share their two cents (like this one), rather than a controversial one

i also click on most links from wikipedia, as any wikipedia entry that gets to the front page seems like it must be very interesting / esoteric. however they turn out generally to be hit or miss

I try not to click on nytimes, washingtonpost and all the corporate/state propaganda junk. And most importantly anything with a question for a headline ( except for Ask HN obviously).

Other than that, I'm open for anything CS/programming/tech/internet related topics.

I hate seeing so much anti-facebook, pro-vegan, climate change, etc obvious agenda pushing by organizations here. Though the mods do a good job of filtering much of it out, but it's obvious there are organizations dedicated to spamming HN to push their agenda.

You rock. This is possibly one of the most interesting questions I periodically ask myself... And never thought of asking the HN community! Bravo.

Click on:

- stuff I know a bit about, and can perhaps offer some anecdata on rather than commentary (because you're all far smarter than me on everything, and that's wonderful) - stuff that reminds me that working for all these hours is worth it after all (because I need it) - stuff that tells me that working for all these hours is preposterous and I should re-assess what I'm doing (because I need this, too) - Ask HN posts that show some human humility and a wanting to learn

Mostly I click on things that teach me something about programming, things that make me feel like my comment will add to the noise, or things that instinctively make me angry.

I avoid everything about startup entrepreneurship like the plague.

- Anything with >500 upvotes that isn't general news, the more technical/esoteric, the better.

- "How to write/build a [something]" if that "something" is fairly low-level, like a compiler.

- Just about anything involving bees.

- Highly-voted threads with mundane headlines -- i.e. either the content is good enough to not need a catchy headline, or its headline has been changed to be less clickbaity -- which might mean the mods thought it was worth saving rather than flagging into oblivion.

- Virtually any Show HNs that happen to make it to the front page

- Ask HNs that have > 50 comments.

I’m a non-technical founder, so I click mostly on:

Accessibility-related articles (my field)

Education/edtech articles (my field)

Most Show HNs that I’m in the target demographic for

Political articles that have lots of upvotes or comments (though I won’t always have the patience for diving into the comments)

Science videos or novelties that appear not to require background knowledge

I rarely click on anything not on the front page from a domain I don’t recognize. Even with all the noscript and ad blockers in the world there are too many baddies out there to justify the risk.

As far as the front page goes I lean toward titles that are about (in no particular order):

Infosec

Politics

Databases

Kernel / OS

News about major tech corps

I upvote stories that I want to see a discussion about or stories that I want to have a discussion about.

I start by scanning through the front page and clicking "HIDE" on:

  Betterridge Headlines
  Articles with a (YEAR)
  Articles with a [PDF] 
  "Do X in Language"
  Blockchain-related
  Mac-related
  "Ask HN: List things"
  Hiring notices

- Specific domains:

  All newspapers
  All longform sites
  All clickbait sites
  All health information
  Medium
  Github
  Wikipedia
  Stratechery

This usually clears out all the time-sinks, ads, "Github spam" and clickbait. And sometimes you find a Tesla article or 2 that's been pushed back to the 2nd or 3rd page.

I tend to (as others have mentioned) focus on languages or fields I'm familiar with. Lots of articles on frontend stuff, and JS libraries, even ones I don't use. Mainly just to get a feel for what is out there. A lot of the time I like to read comments more than the actual article on a lot of things, because I find opinions on HN (specifically comments that have long threads with multiple replies) to offer a better understanding of things than the actual article. A lot of articles are somewhat biased, intentional or not, and reading the comments (most of the time) offers a better perspective from people actually using/working in/experiencing, things the articles are talking about.

I check the comments for any link which looks vaguely interesting. If the top few comments aren't slamming it, I take a closer look.