Ask HN: Do posts by authors with more karma rank higher? As the title states.. Is the HN ranking algorithm based purely on a post/comment's upvotes and engagement, or is the author's karma points an input into the algorithm as well? |
Ask HN: Do posts by authors with more karma rank higher? As the title states.. Is the HN ranking algorithm based purely on a post/comment's upvotes and engagement, or is the author's karma points an input into the algorithm as well? |
The same isn't true of stories I write; I have a pretty good track record of getting blog posts on the front page. But HN doesn't know when something I've submitted is something I've written; what gets stories from high-karma users ranked is mostly just name recognition.
I doubt people are looking at the submitter name when choosing to upvote stories; I do think some subset of them look at the domain name of the story site though.
This might help answer some of your burning questions.
Political news is discouraged by the site guidelines and quickly flagged by users without mods' intervention (source: I'm one of these users). If I want the latest Trump nonsense, I can go to MSNBC or Fox News or Colbert or whatever. I'd rather have more stuff on the front page which we can have a (mostly) constructive conversation about.
and https://news.ycombinator.com/item?id=19212822 in 2019.
Is your perception based on comments appearing near the top right after you post them? Because that's a feature. New comments are always at the top for a few minutes to get their initial set of votes before they move to their natural position in the stack.
Back then I shrugged this of as effect of the voting ring detector but it left me wondering: can the false positive rate be so high and so consistent ... IDK. Needless to say that I never participated in a voting ring and I'm not that well connected that many upvotes should be from the same people.
An alternative explanation that crossed my mind was if it could depend on the story/comment ratio because back at the time I submitted much more stories than I wrote comments. I did this with good intentions, I often like to read what other think even if I don't get involved in the discussion, but I could understand if a platform incentivized a healthy ratio of submissions and discussion.
[1] I can't remember if this was true for stories or comment but I think I could only see it for one type of submission.
The ranking is affected to a good extent by when a post is submitted because of the times (and time zones) when more HN voters are online and active.
I've also posted a link that got buried without anyone looking at it, then came back later in the day to see someone else posted the same thing from a different source and it made the front page.
If there are others gaming their posts here to get karma points they can have them. As far as I know there's no place to cash them in so it doesn't cost me anything.
It's nice to see others find what I've submitted interesting and "karma" gives us a measure of that, but the reason I check into HN is to find things that interest me.
Also, somehow I find myself being disincentivized to post unhelpful comments. As a matter of fact, every time I share an article I found on Hacker News I recommend people reading the comments. This site definitely has the highest helpful to unhelpful comment ratio.
I call it the HN roulette. I think HN is just too large and a lot of good content (or at least, I'd like to think my own content is good, heh) gets missed. Quality of the content is definitely a factor, as are things like an enticing headline, but you also just need to be lucky.
Otherwise, good luck!
20 million karma points and five bucks will buy you a coffee.
(Maybe it would turn into an echo chamber pretty fast, but I'm not 100% certain, which is like I'd like to see someone try it.)
But never seen any evidence that the system gives you preferential treatment if you have good karma.
As much as this probably isn't the fair system we all want, it's what the system's optimization encourages today.
Try posting something SUPER interesting, staying quiet and not telling anyone. It's almost guaranteed not to make it to the front page.
Although I don't encourage this, with the current algorithm, you could probably even prevent others' from getting their work on the front page by posting it ahead of time. When they try to post it they'll get a duplicate link but it's already stale and past its upvote-to-front-page life, which is probably about 30-60 minutes.
Almost everything here is wrong. HN's anti-abuse software isn't perfect, but if you ask friends to upvote your article, there's a high chance it won't help and you'll get your (and their) accounts penalized in the process. We actually go through the penalized posts looking for things to rescue because people tank their own good work in this way so reliably.
The dupe detector does not work the way you described.
If you post something super interesting that doesn't get attention, you (or anyone) can always email us at hn@ycombinator.com and we might put it in the second-chance pool (https://news.ycombinator.com/item?id=11662380), in which case it will get a random placement on HN's front page.
I've seen lots of friends (including dozens of YC founders) ask for upvotes, usually via FB in ways that wouldn't necessarily constitute a ring (e.g. asking strangers in founder-friendly facebook groups), and they're usually of things worthy of publicity, and they do make it to the front page pretty quickly after the upvote requests.
I've also seen a lot of instances where posting once didn't do anything, and then posting a second time with an equivalent but different URL + asking a few friends to help upvote made it to the front page pretty quickly.
For example, I posted https://github.com/dheera/rosshow on 2019-Mar-27 -- no dice.
Then posted it again with a "www.", i.e. https://www.github.com/dheera/rosshow -- then asked a couple friends to upvote -- and then BAM front page (https://news.ycombinator.com/item?id=19519165).
So the only thing I did differently between trial A and B was asked people to upvote, and trial B succeeded.
This is not true. I don't do this and 5/13 of my submissions hit the front page in the last 12 months.
The most important thing is to find an article that already has a suitable title so that you don't have to editorialize it against the site guidelines just to get the crucial first upvotes. It's a shame that often the best article on a topic doesn't have the best title (often when it is a primary source). Sometimes the mods change the link later.
If the article isn't particularly good, and especially if someone is just trying to promote something, woe betide.
I've never asked anyone to boost a post and being a rather nati-social creature I don't have an audience of devoted fans, if anything the opposite. I could get more karma if I was selective about the time of day I posted (eg lunchtimes are usually good) but I just post stuff as I discover it.
To me, it seems obvious that the value of the content and timing are the two biggest levers. I've gone to post things that are new and interesting, but someone else posted it, my submission counted as an upvote, and the submission ended up on the front page.
I have never asked anyone to upvote anything.
But I know for a fact this isn't the _only_ way to reach the front-page of HN because I didn't ask anyone to upvote this post. ¯\_(ツ)_/¯
Ideally, whether or not something makes it to the front page should depend on whether the readers of HN think it's interesting.
However, the current (sub-optimal IMO) situation is that as long as it meets some pretty low bar for "interesting", it depends MUCH more on the submitter's friend network than the readers.
I am often inclined to flag political threads here if they get egregiously bad. It very often gets to be the opposite of interesting.
Despite that I have actually learned stuff here from people that I originally disagreed deeply with.
(Oh, and I am also fairly sure certain people have/could learned/learn something from me, but thats up to them to seize the opportunity. d:-)
> I am often inclined to flag political threads here if they get egregiously bad.
I wish they were better but often I wish people could stick to flagging bad behavior instead of the entire thread.
That said, I am not here to challenge moderator decisions. Even with showdead on this place is fantastic compared to certain places I find it easy to compare with.
Agreed. The incidence rate of people who can intelligently argue a controversial position is much higher here than in the average population.