I used to challenge people with this on Reddit and the results were sometimes silly ridiculous. In an echo chamber like environment people sometimes downvote because they are compelled by the gravity of Reddit’s visible vote count and sometimes because they are willfully using a downvote as a blunt weapon to suppress a challenging opinion. Yet when you compel people to put that into words the results are clearly a form of cognitive complexity with people twisting themselves into knots to describe their own behavior in a meaningful way. That absurdity aside I would notice that the challenge of asking people to qualify their downvote would slow or halt the pace of the echo chamber foolishness.

HN is not Reddit though. The maximum downvote count is 5 reducing a comment to a score of -4 and aside from the color fading of negativity scored comments vote counts are not published. The majority of users and discussions are also generally superior to those of Reddit as well.

Even with that said problematic discussions do happen on HN even if far more rarely. The reason for the problematic behavior, as many people clearly identify, is people looking for agreement more than discussion or insight. In this case people on HN engaging conversations only for agreement are generally eager to advertise their censoring foolishness and insecurity.

Aside from contrived political discussions downvotes are generally reserved for comments that are far off topic or impose violating behavior.

Either way the biggest change this would result is visibly associating a user to their down vote. Otherwise, I don’t see this having a dramatic impact on HN. This would have a huge impact at Reddit though.

Thanks for your detailed reply!

Echo chambers are exactly the thing to avoid, both up and downvoting.

We know that downvoting will suppress activity more than upvoting.

The downvoting limit to -4 is a new to me so thanks for sharing that.

Perhaps the downvote reason could be published anonymously to the original poster and doesn't have to be public... or that could be optional.

An extensive explanation from 'dang, 10 months ago:

https://news.ycombinator.com/item?id=20864278

An excerpt:

> The counterargument is that, just as with downvote-reason-giving, downvote-reason-reviewing would be dominated by the same forces as downvoting in the first place. You wouldn't get more signal, just more complexity, plus a lot more work—both because of the reviewing itself, and because of meta-quarrels about reasons and reason-reviews.

There's been a lot of comments and other Ask HNs on this and related topics in the past. The search box can at the bottom of the page can help you find some of this.

Sweet, thanks!

Part of me wonders if it would be an expensive temporary experiment instead of trying to imagine it.

Downvotes are a valuable signal. Adding that kind of friction means most people won't bother.

Maybe it could work if there were pre-determined answers one could select (so it's faster) but I don't see how the simple UI could display that.

Food for thought: should upvotes require a reason too?

Food for the food for thought: Is more learnt from disagreeing, or agreeing?

I don't think people mind being wrong.

Or maybe people like pointing out others are wrong more than sticking around for the conversation.

We do know the quality and diversity of opinion and conversation doesn't suffer from upvoting, but it can be suppressd by downvoting.

If downvoting had a requirement of a one sentence input, that was maybe anonymized or made only visible to the OP... maybe there's something there.

How many downvotes wouldn't happen if it wasn't quite bad enough to articulate into one sentence?

> Is more learnt from disagreeing, or agreeing?

Your comment reminded me of Polis:

https://pol.is/ https://github.com/pol-is/

Taiwan use it for their multi-stakeholder decision making to find points of agreement. Their application of it to the Uber vs Taxis situation was quite interesting.

https://debconf18.debconf.org/talks/135-q-a-session-with-min... https://blog.pol.is/pol-is-in-taiwan-da7570d372b5 https://blog.pol.is/uber-responds-to-vtaiwans-coherent-blend...

The quality can be affected by poor upvoting. I've seen too many examples where the first comment is incorrect technically.

Adding a reason wouldn't help.

Maybe providing a reason over a certain number of upvotes might be a good idea.

Yeah downvotes would be a great signal if they were public and you could block those guys.

I think a large part of the problem is the ambiguity and subjectivity of what upvote/downvote is supposed to represent. Upvoting can signify intent to promote visibility (intended meaning on HN, I think), agreement with opinions, amusement, compassion, relevance.

Oftentimes these are in conflict. Do I upvote a comment that IMO presents dangerous thoughts built on false assumptions and flawed reasoning, but that has a solid dismissal and spawns a great and honest conversation subthread? When we see highly voted and opinionated comments, I think we all generally believe this is because most voters agreed with the sentiment. I have no issues downvoting useless comments I disagree with, but these are tricky. There are other situations when the mixing signals become an issue.

I've been brooding on how a more multi-faceted voting system would work - how many dimensions are there, how are they presented and how do they affect thread rendering? What's the right balance and how specific should it be for different communities (think subreddits)? My hunch is that 1 (status quo) is too little and 5 too many.

Slashdot did a version of this or maybe still does. It’s time consuming. Also people tend to vote ideologically and down vote if they disagree even if you have something interesting to say.

There are a lot of topics that people passionately argue in which they are not expert and it’s disheartening. It’s News for Nerds but we can be a fickle bunch. I think the posts that are tech go better than others.

I think it might do even more good to leave the system as-is, but add a short pop up whenever you upvote or downvote (clickthrough, lasts 1 second) that says “this made me think” or “this doesn’t promote discussion.” That sort of cue would subtly encourage good culture on a forum and it’s the sort of thing I plan to include in my own systems one day (if it comes up)

Love this idea!

Curious why the downvotes.

Maybe someone trying to be epic, or funny