It's been interesting to watch the rise and fall of internet comment sections. They used to be found at the bottom of just about every page. But the moderation burden has led to their removal in more and more use cases.
News sites started stripping comments from articles so long ago. When you see a comment section on the bottom of an article or blog, it contributes to that feeling that the website hasn't had a remodel in a long time. Like an animated gif spacer from the 90's web.
Now we are seeing the social media hubs, technically places that are entirely comment threads, abdicating moderation responsibility to thread creators. Twitter users can hide replies to their tweets. Youtube creators can as well. Now this.
Overall a good trend, IMO. Moderation responsibility was so diffused that no one did it well.
The emerging pattern is that the site owner can provide global moderation of clear cut illegal activity. This doesn't meet anyone's bar for civil discourse, and there's not a good global solution for how to achieve that, so the site owners will place the next level of moderation in users' hands. This starts to model how it works in the physical world, so I think we're on the right track.
It also highlights how valuable communities that do have good moderation can be. HN being the major example. I would probably include some (emphasis on the "some") subreddits.
I assign very little value to anything on the web that can't be commented on. Sure, a solid piece of journalism does have value but more so if people can add to the conversation.
Users should be able to moderate though and moderation points should be earned and spent and provide some context. Users should be able to browse comments based on their preferences, revealing those posts which have been down-voted out of sight, or made anonymously, if they choose.
Everything on the web can be commented on. Post the URL to Reddit, Twitter, wherever. Add your comments underneath. There you go, conversation started.
That doesn't mean that the publisher has to host your opinions on their site. It's their site, they choose what to put on it. They're under no obligation to amplify your opinions.
Your first and second sentence contradict one another. You assign 'very little value to anything that can't be commented on', yet you believe a 'solid piece of journalism does have value', even if it can't be commented on.
I would say the overwhelming majority of quality writing on the internet does not accommodate comments, from academic journals, to serious think-pieces and long-form essays.
"Letters to the Editor" used to (and I suppose still does) function as a method of replying to serious think pieces. Of course the newspaper/magazine curated (and edited) which replies were printed and just as importantly there's a built in speed limit to that kind of exchange. Both these things help (although you do still see incredibly stupid things in letters to the editor).
I would disagree, in that a presumptively factual news article should not also include reader opinion as context. That seems counter productive to news distribution. I would suggest that we have a opinion crisis in journalism already, and comments only contribute to the problem.
User point systems are very lopsided to users that have the time to generate points, which is probably a negative selector for quality/experience/diversity in opinion.
I still look for comments on any article I find interesting. If the site owner takes the time to moderate the comments you can usually find some really good info there. I like to offer comments on my own sites because it's an easy way for readers to give me feedback and to start a discussion.
They're a great barometer of the quality of the content and dedication of the team running the website to said quality. Its not surprising in the least that Facebook falls short in this metric.
This is becoming more and more futile as time goes on. 10+ years ago blogs were fun because of the comment section. Each website had its own particular audience who created their own culture in the comments.
Now that's pretty much all gone. Most 'blogs' have no personality and are producing the same kind of keyword-researched content every other site is. The writing is dry and lifeless, as if its only purpose is to be picked up by a search engine. Scroll to the bottom, no comments, just a grotesque "share" widget.
These sites look and feel like ghost towns now. It's a completely different atmosphere than what internet communities were envisioned to be.
For some websites I like reading comment sections too.
Perhaps a good moderation strategy would be to have a minimum length rule for each comment? It would not prevent spammers and bots, however putting those aside only people who want to deliver a thought-out message would participate. Not saying there wouldn't be controversial messages though.
Maybe animated gif horizontal rule is more accurate. One that comes to mind is a lightsaber, laying on its side. Placed between paragraphs (hopefully) about Star Wars.
Alternatively, page visitor counters would have been another example of outdated web aesthetics.
> News sites started stripping comments from articles so long ago. When you see a comment section on the bottom of an article or blog, it contributes to that feeling that the website hasn't had a remodel in a long time.
I've noticed that this is correlated with AMP adoption. Many news sites still have comments on the "full" page but not the AMP page.
"Comments", free-floating on the bottom of news articles, end up being pretty much the opposite of a community. They're a venue for long running partisan fights. People running pages have realized that they're the equivalent of putting up a sign saying "please tell me I suck" in public - everyone with any kind of complaint is all over them.
It's increasingly clear that if you want to have a community on the internet, rather than just host a warzone on your lawn, it has to have defense mechanisms in place. This post from Doreen the other day resonated with me: https://news.ycombinator.com/item?id=26639182
For user: the story feature is kinda like blog. They can share their thoughts, images, videos, stuff they are interested in with few clicks. General public never owned anything like blogs because it is not easy to set up as sharing stories. Also, stories actually engages followers/contacts with the content. I think that is why it's popular among users.
For companies: Best place to put ads is between stories. You get 100% views for the ads and hence more engagement. That is why every company is running after it.
Yeah but it appears to be a bit more specialized and narrowly focused than a blog because a) it's apparently deleted after a few minutes or really hard to find after a short interval, and b) responses and reactions are shown only to that person and not the others who view ie. Also I don't know what makes this more true for the story than the feed: "stories actually engages followers/contacts with the content."
Interesting idea. It'd likely kill off trolling or low quality comments as there wouldn't be the eyeballs for engagement for dopamine hits to reinforce that level of behaviour.
A similar approach could work in "hierarchical/threaded" forums: comments are private between parent and child, unless the parent says the child comment can be public.
Yes, that's a nice moderation approach - allowing OP to be curator. It also allows OP the opportunity to comment before the thread starts to become public - giving a chance for a response to clarify, rebut, or other.
Think about how that would work in reverse. Lets say you post something fake, like 5G causes covid. And any response that calls out your post as being fake, you just hide.
Which is why you need a "God" level of moderation (or really just parenting, parents watching the child in the playground as we all try to understand and master consciousness) for every platform - where they have the ability to add a warning tag/comment - or moderate that person off of platform either as a timeout or permanently; and then having many platforms that are privately owned deciding how they govern is how we de-risk government tyrannical control mechanisms being in place that then apply to all platforms.
And then users decide/"vote" with what platform or network they align with based on if they use it or not; the issue of allowing users and their networks to being easily mobile/data and network operability is important then to reduce the friction of this being possible.
And then the layers of infrastructure are additional layers of decentralized control mechanisms, e.g. Amazon/AWS kicking off Parler, Parler then needing to eventually find infrastructure that was supportive of their lack of moderation against calls for violence, etc. At least then lines are drawn and we can see who's allying with who, and people can respond or behave accordingly.
> They're a venue for long running partisan fights.
Worse, they tend to be completely uninformed emotional ramblings. I often wonder just how many of those comments and votes are from bots that only exist to provoke gullible people.
In my experience, people who use "emotional" in a derogatory way are often the most emotional themselves because they're blind to their own feelings. Being emotional is not bad.
I didn't use emotional in a derogatory way, but I did use it to create a loaded statement. The implication of the word emotional in that statement is meaningless unless somebody reading the hypothetical comment is emotionally activated.
No. It's just used that way by people who are getting fed up with reading:
1. made up facts from organisations trying to incite emotional responses rather than conveying accurate information,
2. comments from people who are ignorant on specific topics but still feel strongly enough about said topic to voice their opinion.
Clearly we are never going to remove emotion from content. That's as unrealistic as asking the news to be unbiased. But wouldn't it be nice if there was a little less emotive bullshit being argued as factual information....
Unfortunately turning off comments on select posts isn't going to fix the rot.
Just read this yesterday from Fred Wilson talking about Tumblr founder resisting the idea of comments. Only thumbs up and reposts were allowed. The founder said it was a wasteland for negativity.
Have there been any studies about the "optimal" size of a community? It feels like there's a critical mass so long as the size of a community remains less than n people it remains civil and people can forge actual relationships (even behind "anonymous" usernames). Reddit and Discord come to mind. The smaller subs and servers are great while the huge subs and servers tend to be toxic internet wastelands.
I think a good way to measure this effect is Discord. You have servers of varying sizes and can compare relative “toxicity” between them. IMO things seem to get pretty bad after 100-150 people.
Personally, I've always preferred moderation controlled by the client (i.e., a killfile in a NNTP client) rather than relying on moderators. That way, I can decide what I see.
Theoretically, bad actors would end up in everyone else's killfile, but in reality, it never really happened. Some clients at least allowed hiding a subthread so that you wouldn't see messages that were essentially a dispute between two people.
But one way to slow down the flame wars would be to rate limit the number of comments that can be posted. If you require people to wait 10 minutes to post a subsequent comment, that would definitely slow down the discussion and probably would lead to them moving on to something else.
The next stage is to let people remove likes and hide share counts. I think we have systems which pay way too much attention to these numbers, considering how trival they are to fake.
Correct. Instagram is hiding “true” like counts for everyone but the poster. So, if 2500 people like a post, it’ll just say “Liked by ${RANDOMUSER} and thousands of others”. The full list of people who liked is still available to everyone, just not the exact number (without manual counting (again, unless you’re the poster)).
Gawker/Kinja - and specifically Jalopnik.com - seemed to have figured out how to build community in the comments. They simply have the author review them in batch form before they make it to the public.
Nearly. Not "really." The tragedy of the commons is not solvable. I accept that, and slashdot made it trivial to read without seeing those shit posts. It was probably just filling up their database with racist and faciast novels.
Jalopnik has it's fair share of memes and repetitive in-jokes (BMW blinkers, Mustangs hopping curbs). But there are are also some pretty insightful and knowledgeable threads. I've seen a couple of technical comments on certain vehicles that I'm absolutely convinced were written by an engineer that designed that vehicle.
The only comments section I have recently seen that were filled with thought provoking, insightful comments and/or constructive criticism were the New York Times and the Athletic. It's no coincidence that both require me to pay to subscribe and get access to them.
I'd like to see even more fine grained controls for comments on posts for social media. Facebook should emulate what Twitter has done and let me maintain an "Allow List" of friends that are allowed to comment on my posts, links, etc. The friends system was originally designed the way it was because you weren't supposed to open your Facebook friends to everyone, but 15+ years and the shifting of connections that time brings (good friends may now just be acquaintances) makes this feature a necessity.
Facebook is already an echo chamber. How much would that amplify things?
Hypothetically, aunt Joan just shared the post about the Ever Given ship containing tons of children in cages, and it's all Hillary Clinton's fault. Since she only allows her conspiracy theory believing friends to comment on her posts, there's no disagreement about the cause of the Suez Canal blockage. It's accepted as fact among anyone who can comment, and it's uncontested for the ones who could be swayed.
The Aunt Joans of the world are often presented as the problem here but mainstream news outlets like the New York Times and the Atlantic that have promoted hoaxes like the Iraqi WMD deception and Russiagate hysteria have done immeasurably more damage.
This already exists. Remove them from being connections. It's exactly the same outcome - what do you think these people will think when they see one of your posts and notice you've barred them from commenting?
I would just remove the connection instantly in that scenario.
Well, it might be a mistake, but I find that the best communities I'm a part of are echo chambers.
I like urbanism Facebook groups where we can discuss how to improve a specific intersection, for example. Where we are like-minded people and we try to find the best solution.
When we have the same discussion in the neighborhood group, the thread becomes a shouting match with drivers requesting to be able to run through the intersection at full speed with no inconvenience and to get cyclists and pedestrians out of the way. With the usual "they don't pay taxes". So it is less interesting.
63 comments
[ 3.1 ms ] story [ 116 ms ] threadNews sites started stripping comments from articles so long ago. When you see a comment section on the bottom of an article or blog, it contributes to that feeling that the website hasn't had a remodel in a long time. Like an animated gif spacer from the 90's web.
Now we are seeing the social media hubs, technically places that are entirely comment threads, abdicating moderation responsibility to thread creators. Twitter users can hide replies to their tweets. Youtube creators can as well. Now this.
Overall a good trend, IMO. Moderation responsibility was so diffused that no one did it well.
The emerging pattern is that the site owner can provide global moderation of clear cut illegal activity. This doesn't meet anyone's bar for civil discourse, and there's not a good global solution for how to achieve that, so the site owners will place the next level of moderation in users' hands. This starts to model how it works in the physical world, so I think we're on the right track.
Anywhere else?
Users should be able to moderate though and moderation points should be earned and spent and provide some context. Users should be able to browse comments based on their preferences, revealing those posts which have been down-voted out of sight, or made anonymously, if they choose.
That doesn't mean that the publisher has to host your opinions on their site. It's their site, they choose what to put on it. They're under no obligation to amplify your opinions.
I would say the overwhelming majority of quality writing on the internet does not accommodate comments, from academic journals, to serious think-pieces and long-form essays.
User point systems are very lopsided to users that have the time to generate points, which is probably a negative selector for quality/experience/diversity in opinion.
Now that's pretty much all gone. Most 'blogs' have no personality and are producing the same kind of keyword-researched content every other site is. The writing is dry and lifeless, as if its only purpose is to be picked up by a search engine. Scroll to the bottom, no comments, just a grotesque "share" widget.
These sites look and feel like ghost towns now. It's a completely different atmosphere than what internet communities were envisioned to be.
Perhaps a good moderation strategy would be to have a minimum length rule for each comment? It would not prevent spammers and bots, however putting those aside only people who want to deliver a thought-out message would participate. Not saying there wouldn't be controversial messages though.
I know what a “GIF spacer” is (a 1x1 pixel image resized to the desired width and height), but what’s an “animated GIF spacer”?
https://www.oocities.org/itsamemario2000/english.html?202025 the animated green lines.
Alternatively, page visitor counters would have been another example of outdated web aesthetics.
I've noticed that this is correlated with AMP adoption. Many news sites still have comments on the "full" page but not the AMP page.
It's increasingly clear that if you want to have a community on the internet, rather than just host a warzone on your lawn, it has to have defense mechanisms in place. This post from Doreen the other day resonated with me: https://news.ycombinator.com/item?id=26639182
For companies: Best place to put ads is between stories. You get 100% views for the ads and hence more engagement. That is why every company is running after it.
And then users decide/"vote" with what platform or network they align with based on if they use it or not; the issue of allowing users and their networks to being easily mobile/data and network operability is important then to reduce the friction of this being possible.
And then the layers of infrastructure are additional layers of decentralized control mechanisms, e.g. Amazon/AWS kicking off Parler, Parler then needing to eventually find infrastructure that was supportive of their lack of moderation against calls for violence, etc. At least then lines are drawn and we can see who's allying with who, and people can respond or behave accordingly.
So if this happened on HN, the subthread probably wouldn't survive long.
Worse, they tend to be completely uninformed emotional ramblings. I often wonder just how many of those comments and votes are from bots that only exist to provoke gullible people.
Do you really think acting like an emotionless robot is a good way to communicate with others?
1. made up facts from organisations trying to incite emotional responses rather than conveying accurate information,
2. comments from people who are ignorant on specific topics but still feel strongly enough about said topic to voice their opinion.
Clearly we are never going to remove emotion from content. That's as unrealistic as asking the news to be unbiased. But wouldn't it be nice if there was a little less emotive bullshit being argued as factual information....
Unfortunately turning off comments on select posts isn't going to fix the rot.
https://avc.com/2019/08/tumblr-2/
https://www.reddit.com/r/somethingimade/
Once it exceeds Dunbar's number you are getting different tribes fighting each other, that's just the human nature.
Theoretically, bad actors would end up in everyone else's killfile, but in reality, it never really happened. Some clients at least allowed hiding a subthread so that you wouldn't see messages that were essentially a dispute between two people.
But one way to slow down the flame wars would be to rate limit the number of comments that can be posted. If you require people to wait 10 minutes to post a subsequent comment, that would definitely slow down the discussion and probably would lead to them moving on to something else.
I'd like to see even more fine grained controls for comments on posts for social media. Facebook should emulate what Twitter has done and let me maintain an "Allow List" of friends that are allowed to comment on my posts, links, etc. The friends system was originally designed the way it was because you weren't supposed to open your Facebook friends to everyone, but 15+ years and the shifting of connections that time brings (good friends may now just be acquaintances) makes this feature a necessity.
I like that.
Hypothetically, aunt Joan just shared the post about the Ever Given ship containing tons of children in cages, and it's all Hillary Clinton's fault. Since she only allows her conspiracy theory believing friends to comment on her posts, there's no disagreement about the cause of the Suez Canal blockage. It's accepted as fact among anyone who can comment, and it's uncontested for the ones who could be swayed.
I would just remove the connection instantly in that scenario.
I like urbanism Facebook groups where we can discuss how to improve a specific intersection, for example. Where we are like-minded people and we try to find the best solution.
When we have the same discussion in the neighborhood group, the thread becomes a shouting match with drivers requesting to be able to run through the intersection at full speed with no inconvenience and to get cyclists and pedestrians out of the way. With the usual "they don't pay taxes". So it is less interesting.
Just how many missteps can one company make?