Ask HN: What's the Secret to Ranking on HN?

https://news.ycombinator.com/item?id=39030185

by adaboese • 3 years ago

2 32 3 years ago

news.ycombinator.com

I thought I wrote a really interesting article and shared to HN (Using Vector Embeddings to Overengineer 404 pages), but it got 0 upvotes.

What am I doing wrong? Is the article not as interesting as I thought it is? Is HN not my audience?

Just looking for advice for future posts.

gigatexal• 3 years ago

Periodically reshare it. It could be time of day or really anything.

mtmail• 3 years ago

Not enough users browse the https://news.ycombinator.com/newest queue, there's 2 submissions per minute. I've seen articles ignored and then suddenly the third or fifth time it hits number 1 on the frontpage. It's much more random than you might expect.

adaboese11• 3 years ago

so re-submitting after a couple of hours is really the way to go?

ThrowawayR22• 3 years ago

Maybe once at most. Re-submitting more than once within a few days will likely get it flagged by users.

PaulHoule5• 3 years ago

I have statistical models that can predict if a headline will get > 10 votes or if a headline that gets > 10 votes will get a comments/votes ratio > 0.5. (We can’t think of a better headline than “Richard Stallman is dead”)

Both of those are treated as probability calibrated classifiers because of the extreme variations in outcomes. You might post something one day and get 0 upvotes, another day somebody might post it and get 500. Not that many people read the “new” page so it is really iffy if you get your first few votes.

In principle you could get more information out of the data treating it as a regression problem (e.g. give more points for predicting that an article gets 500 votes as opposed to 50) but one model I made never predicted that anything gets more than 16.7 votes which means the error function is going to always look awful and not really be a good tool for improving the model. The logit models I”m using, however, give realistic scores in the face of uncertainty.

adaboese11• 3 years ago

How does that model rank the title I shared?

pdyc3• 3 years ago

interesting, can you run it on my blog? i bet it would generate false negative for you.

jacquesm• 3 years ago

You're spamming SEO junk, please don't.

adaboese11• 3 years ago

In what way on Earth is the post that I am referring to an SEO junk?

KomoD5• 3 years ago

Your entire business looks to me like SEO spam, specifically about generating it

ThrowawayR22• 3 years ago

Why would you care about trying to forcibly increase ranking on HN of an otherwise unremarkable post if you aren't doing SEO or advertising your business, neither of which is welcome here?

KomoD5• 3 years ago

Post things people are interested in

pdyc3• 3 years ago

that is necessary but not sufficient. I too find ranking to be unpredictable.

mindcrime2• 3 years ago

What am I doing wrong?

Probably nothing.

Is the article not as interesting as I thought it is?

Didn't read the article[1], but the title sounds interesting enough.

Is HN not my audience?

I don't think it's that. As others have said, the /new page gets a LOT of traffic, and not that many people read it and upvote stuff. So there's a lot of competition for not so much attention. Given that, luck is going to play a big factor. If somebody who's particularly interested in vector embeddings happens to see your post and upvotes it, that can trigger a cascade where it gets a bunch of upvotes. If they don't, it can remain at 0 upvotes forever until it vanishes off the first page of /new, where it's (much) less likely to ever be seen.

I wouldn't worry about it. Other than to say, if your marketing strategy depends on getting stories on the HN front page all the time, or on any kind of predictable basis, you're probably barking up the wrong tree. Submit stuff that you think is interesting and just let it be. Don't try to "game the system" or you'll wind up shadow-banned (either your user account, or your domain).

Note the following blurb from the HN Guidelines[2]:

Please don't use HN primarily for promotion. It's ok to post your own stuff part of the time, but the primary use of the site should be for curiosity.

[1] https://aimd.app/blog/2024-01-16-using-ai-to-overengineer-40...

[2] https://news.ycombinator.com/newsguidelines.html

Ask HN: What's the Secret to Ranking on HN?

Related Stories