Twitter without downvotes is, after all, famously not a single train of thought under a Tweet!

But seriously: HN has downvotes, and almost every thread follows the basic structure of a somewhat healthy argumentative environment (alternating positive-negative comments down each child node). Things get downvoted (and you're entitled to any number of sour feelings about which things those are), but the overall structure of this site's comments demonstrates that up/downvotes aren't themselves toxic to discussion.

> Twitter without downvotes is, after all, famously not a single train of thought under a Tweet!

In my experience, it really isn't! Whenever I see a bad take on Twitter, I generally see some sort of rebuttal within the first dozen or so replies. If you expand that rebuttal, you start to see some argumentation, which to me is the interesting part of Twitter.

I think part of what makes HN's downvote system viable is that it's a well-defined community, you need a lot of karma relatively speaking to unlock the downvote privilege, and in practice comments are only downvoted if they are off-topic or poorly articulated.

In that same vein Stack Exchange's downvote system works well too, because karma (rep) is tracked individually per-community and you likewise need a certain amount of it to unlock the downvote privilege. This makes it so that on SE and HN alike, the people with some seniority in the community are the ones to help dictate what should be surfaced by the site.

On Twitter though, it's liable to turn into the same mess Reddit has. Because there is no hard line between different communities, you can't have a meaningful per-community karma limit for the downvote privilege. This eventually devolves into a free-for-all where users brigade other subreddits (and on Twitter, other users' feeds) to mass downvote opinions they disagree with, no matter how articulate or well-sourced they are.

> it's liable to turn into the same mess Reddit has

Reddit has its problems but the idea that it’s somehow worse than Twitter is laughable.

Even with vote brigading subreddits are moderated and so actual content brigading is much rarer. Can you say the same about Twitter?

To me they have the ability to be equally as bad (and this is coming from someone who see’s the negative view of social media as overblown). What makes Reddit a bit more dangerous is their higher technical ability to do damage as opposed to Twitter social ability. In my opinion Reddit is more susceptible to hive mind theatrics since the site is specific designed with communities/groups in mind. You can bookmark a subreddit page and spend your entire experience on that subreddit. Twitter makes that a bit difficult as, while you can choose who you can follow, you cannot choose what the people you follow tweet, retweet, reply to which will inevitably expose you to different views and vice versa.

You can try to build a bubble on Twitter but something will inevitably slip in.

Maybe we're in different bubbles, but my experience with bad takes on Twitter is that the entire first level is mockery and quote-tweets, with people doubling down on that mockery in the responses to the quote-tweets. Rebuttals, if they come, happen mostly on locked down accounts or via subtweeting (although mockery is also standard in the subtweets).

I also don't think that up/downvotes are likely to change that for Twitter, since Twitter's culture is essentially built around bull sessions[1]. A key part of socializing during bull sessions is mockery, and getting buy-in from your group to perform mockery/ragging on another person in then group (who is supposed to gracefully accept said mockery).

[1]: https://www.urbandictionary.com/define.php?term=bull%20sessi...

HN also has moderators who can step in when things look like they're getting out of control which really helps. Or in some cases just point people to the rules and tell them to stop causing flamewars or get out.

It really makes a difference to the overall discourse. Something that is missing on most other places on the web!

HN moderation for those of us who have been here longer than some posters have been alive, this place is a lot different than it used to be, and moderation now tends to stifle conversations that in the past would have spawned great conversations that remained civil. Dang and friends do their best but for the most part the community is better off self censoring and exploring ideas and concepts without guidance and supervision. Unfortunately as more people flock to a gold rush a tragedy of the commons occurs and the site starts to see dang and team as absolutely necessary for each discussion when frankly they shouldn't be obviously seen guiding discussion.

The quality of the downvotes depends on the quality of the community, though, and the HN community is a lot higher quality than Twitter.

The size of community contributes also. Twitter has a very broad and huge community comparing to HN one.

HN restricts down voting to users above a certain point count, which really improves the quality of that signal

I'm a bit passionate about this, so my words below may sound "absolutist", but pls understand that being open-minded is really important to me. I'm willing to listen. With that said:

> I really don’t like the idea of downvotes, just makes it easier to make sure there’s only one train of thought under a tweet.

Yes, that's for sure going to happen, and the repercussions need to be dealt with thoughtfully.

But we must do it because there are very few human social situations (in person) where some reflexive negative inhibition is not present, as appropriate reaction. (Maybe therapy, but that's a really specific edge-case.) We give negative feedback all the time in our on-the-fly facial expressions and tones, and most humans have some baked in moderating reactions to that.

We can't have digital spaces without some form of negative feedback. Imagine how fucked up your in-person social groups/interactions would be if everyone, for YEARS was for some reason only capable of boosting/encouraging one another, with no reflexive negative responses -- no frowns, no eyebrow emoting, no body language.

Yes, it will be abused. But we desperately need to figure it out. The relational tools and mechanisms that human brains and bodies have learned over millenia of evolution... these are not things to be discarded so lightly as we have when building tech. They're mostly there for a reason, because they helped us get here as collective. There is wisdom in recreating their patterns. <3

> Imagine how fucked up your in-person social groups/interactions would be if everyone, for YEARS was for some reason only capable of boosting/encouraging one another, with no reflexive negative responses -- no frowns, no eyebrow emoting, no body language.

I think you are really overstating how bad the interactions would be. Phone calls show no body language, and though it might hinder a deeper intuitive understanding of someone's emotional responses, I would say that phone communication is far from "fucked up".

When someone disagrees with what I am saying over a phone call, they can verbally express their disapproval if no body language is available. In fact, verbal disapproval is preferred so that they can express nuance.

When someone disagrees with what I am saying on the internet, they can express their disapproval in a reply if no downvotes are available. In fact, a comment reply is preferred so that they can express nuance.

For many millennia, there has been understandable evolutionary pressure to ostracize people who look and act different from those in your hunter-gatherer tribe, as you may not be able to trust them with your tools, food, and other tribe members. That evolutionary pressure is antithetical to the optimal functioning of a modern globalized society comprised of law and social safety nets, and valuing freedom of expression.

The most important signal nowadays is whether information is well-researched, not whether people agree with it. If not kept tightly in check, voting systems tend to devolve into surfacing the most popular opinions rather than the most useful or even the most true.

For example, some of the most important information I read on the internet fifteen years ago was the stuff that was heavily downvoted back then, in particular, controversial fringe stuff related to medicine and finance. Fifteen years later it's finally entering the mainstream and being upvoted, but I've had access to the information this whole time in small part because I set Reddit not to collapse downvoted comments by default.

This all sounds great if it wasn't for one thing: upvotes. Because of that, we have a system that changes what/when people see things. If we had only replies in order of posting that would be like this suggestion. It tends not to scale when there are very many (some duplicate) comments.

This is exactly what can occasionally be seen here on HN, although I believe the mods here do an incredible job of tailoring the ranking algorithm to not magnify the problem too much.

In many ways I think maybe HN is an example of how you can make the comment down vote work.

Reddit is another example, but it's hit and miss if a subreddit is good quality or not.

The difference between HN and Reddit to me is that HN has consistently higher quality comments.

Will be interesting to see how down votes will affect Twitter.

I don't have a problem with down votes as long as they have zero effect on the conversation thread, unlike here on HN. Greying out and/or making a comment dead does not achieve anything useful. Flagging comments for violating HN terms of service should be more than sufficient to handle things that go out of bounds.

More echo chambers that we don’t need and no one is asking for.

On the contrary, many people do want an echo chamber. I find the key to enjoying my Twitter experience is to differentiate between the threads where people do want a discussion and the ones where they only want to signal how much they agree with each other. Sometimes it's obvious, sometimes not.

No one asks for "echo chambers", but people do ask for a space where they can interact with like minded individuals and push out "trolls".

I want more spaces that push out trolls, but where I can interact with non-like-minded individuals.

If I only interact with like-minded individuals, that means the internet has failed its use case for me, which is to broaden my world view by seeking out opinions that oppose my own.

Of course if by "like-minded" you mean "polite and rule-abiding," then I agree.

That's why I put "trolls" in quotes. The amount of people who self-sort themselves into echo chambers is far more than the people who would freely admit they want to be in an echo chamber.

That's one use case, but not every platform or forum can or should be expected to traffic in free-form controversy. If I'm on a car forum, I expect an "echo chamber" about cars with individuals interested in discussing cars, not to have a confrontational debate about race science or vaccines or Christian apologetics, at least not in the car threads.

Downvotes are tiny murders.