Talk: An open-source commenting platform focused on better conversation
coralproject.net
coralproject.net
And I would be the last to argue that comment sections below news articles are anything but a cesspool - they're notoriously bad! But I'm just not sure a control model this heavy handed and lip-servicey is the right approach.
[1] https://coralproject.net/blog/five-myths-of-community-design...
Of course this is just a theory; until someone implements it and evaluates the result, I may well be completely wrong and maybe it'd even have the oposite effect.
If facebook is any indication, these end up being used mostly ironically. I can't imagine they give facebook any real indication of emotion on their platform.
Even the "angry" react (which is mostly an ironic reaction), is missing the context. Are you angry with the commenter? Are you angry about the something the commenter is referencing?
Interestingly on my site we have a total of 31 custom reactions people can use to express their reaction to a piece. The ONLY one people feel the need to qualify is the dislike one, anytime someone uses it they qualify it with 'disliking the content, not the post'.
There's no obvious reason for it though, the dislike is just one among many and it looks the same as all the other reactions, but it's the only one to generate a unique response.
Perhaps you can attach a react onto the username instead of the comment? Or you can attach a react onto a specific sentence of a comment instead of the entire comment?
Slashdot does that. To vote for a post you have to pick an adjective. I always thought it worked well, because it showed what kind of content would be valued. For example "funny" is an adjective you can pick, which signals the site valued funny comments. I had some great laughs reading it back in the day.
Why not just have an Overton window slider that lets people place where they think the comment is on the scale?
Is there any consensus that a popular website's comment area has high-quality conversation but also does not have heavy-handed moderation?
The Coral website doesn't mention it explicitly but I believe the project is trying to reverse the trend of publishing websites disabling user comments.[0][1]
It seems like a bunch of websites suddenly realized their user comments were more of a liability than an asset. Either heavily moderate them (costs $$$ in human labor) -- or disable them completely.
[0] https://www.wired.com/2015/10/brief-history-of-the-demise-of...
[1] https://www.kqed.org/lowdown/29720/no-comment-why-a-growing-...
But from my experience there is situations and places where someone has to look for people and help them to stop their destructive behavior.
In my opinion this problem is connected to the "tragedy of the commons" and maybe some day we (humans) will overcome these flaws but until then they need some leadership, control and sanctions (just like kids and teenagers when they grow up).
Of course there is people who can behave all time when they communicate and who don't leave their trash everywhere they go but I think it's a minority.
While saying this I'm very aware of the fact that I wrote a lot of bad comments myself. Maybe such a system could help people like me, I'd be glad to see it in action to find out.
4chan solved this problem a decade ago. Put your content before your identity rather than the other way around.
In many spaces today, being a straight white man is far more inflammatory than being a non-binary person of color. This is especially true in the open-source community.
The solution is to stop pushing your identity onto people when it's completely irrelevant. I'm not saying hide who you are. But it's simply not relevant to the vast majority of discussions.
EDIT: OP literally says: “ I'm not saying hide who you are. But it's simply not relevant to the vast majority of discussions.”
Again, who gets to make that decision? Who gets to set the criteria for making that decision?
The people you're talking to. These are all just social mores, but you are the one that deals with the fallout.
There's pushback on identity first discourse because it's often seen as distasteful.
Much like what Instagram is trying now (hiding other people's likes count) or this very platform, doing away with those metrics should help foster better conversations.
The concern mentioned in the excerpt from TFA was a perception by the people it spoke to that they had to self-censor in order not to be personally attacked in online discussion.
You make a good point but it seems like it is about something other than what the TFA excerpt describes.
TFA quoted by OP:
> participants talked about having to constantly run a cost-benefit analysis on participation in any online conversations, based on the likelihood that they would be attacked, merely for participation.
But it's also extremely egalitarian. Everyone is on the same, whether you be an old* or a new*.
I don't partake but I like to dive in and observe them. Fascinating, because as much as we'd like to deny it, 4chan is the whole spectrum of humanity.
Personally I think all content should be unfiltered and the people who cannot handle this, should learn it. Open discussion is difficult when you have to self-censor all the time. We have it so bad, there's even a bunch of words we're forbidden from using in filtered communities and it's sold as a good thing. Lots of ideas and opinions are also banned, which is really unfortunate. We tend to criticize China for exactly the same things that are completely normal for most of us. It's just that China bans different ideas.
Having a high tolerance for vulgarity and dank Nazi memes is not the same as having open discussion or tolerance for opposing points of view. People who rally behind sites like 4chan, Voat and Gab in the name of "free speech" just want to be able to shitpost hate in comfort without consequence - they want their own filter, not freedom, certainly not for anyone who disagrees with them.
I think it's a perfectly valid opinion on my part to consider this garbage and to not want its stench everywhere.
Not because I'm afraid to have my views challenged, or to be insulted, but because it shouldn't be my responsibility to have to turn every racist and bigot on the internet, particularly when opinions to the contrary abound and they could just as well educate themselves, and because it gets tiring to deal with this kind of nonsense all the time.
This isn't even true in my experience. Frankly, it's a reputation that helps to keep the riff raff out of the space. Chan communities are the best kept secret on the internet, in my opinion.
No, seriously, if you're a person who is not measuring what they're saying to see what the cost may be, you can expect attacks to come your way for your own carelessness, and it has nothing to do with your identity. Being able to gush at the mouth is not, in fact, a sign of privilege, but is rather more rope to hang yourself with.
Doesn’t really sound like a thriving wellspring of diverse opinions.
And I don't think you need to be from a diverse set of locations to have a diverse set of opinions.
“People who are different from one another in race, gender and other dimensions bring unique information and experiences [...]”
https://www.scientificamerican.com/article/how-diversity-mak...
As for the the actual article you linked, the author herself has nice race-bait titles of work such as "The White standard: Racial bias in leader categorization", you can't seriously tell me that this is an objective body of work? There is also the piece she wrote called "Office Holiday Parties Highlight Racial Dissimilarities and Fail to Promote Team Unity" ... apparently the office party isn't safe from race based politics. Sorry but this sets of alarm bells in my head.
The piece itself doesn't cite any evidence for anything it claims.
People would be surprised.
I was on 4chan many years ago, I went to the Anonymous protests in London. I have thick skin and I know not to take stuff seriously, but in the last few years especially it seems like it's become dominated by /pol/ and people who are legitimately white supremacist traditionalist alt-right scum, because those communities seized upon the edgy culture of /pol/ as it used to be and use it as a channel to push their very real non-ironic toxic politics.
The other explanation is "those people are mean and I don't want to hang around mean people", but that's really not all that different from the first explanation.
https://www.youtube.com/watch?v=vw9zyxm860Q
Obviously this is trolling, but it damn impressive they found it.
They've done quite a lot of other stuff including finding ISIS targets and getting them killed, using nothing bit a sheet of paper and an innocuous message to make the media look silly, they've even caught animal abusers.
Plenty of people stick their nose up to them but they are probably one of the more effective online communities. It just depends whether you happen to like what they are effective at.
As a straight white man, this is bullshit. I have many close friends who are part of the LGBT community, so I can see their experiences compared to my own. They don't go around "pushing their identity onto people" and neither do I. We're all just normal people trying to build cool things.
And I see my experiences are much more positive than theirs.
Preface a statement with "as a non-binary person of color..." and you get applause.
The trick is: don't be an asshole and recognize that other people have different experiences which are just as valid as your own. Every time I've seen someone talking about how it's terrible to be a white man these days, it's because they want to be an asshole to someone and are being told that's wrong. Straight white men who are actually kind and listen to others, seriously considering their opinions don't get much flak for being straight and white. Is this really that difficult to understand?
https://github.com/coralproject/talk/graphs/contributors
Why is that? (In the summer of 2018 and onwards, 90%? 95%? of the GitHub activity abruptly disappeared)
Two other open source blog commenting system that are being actively developed are Commento https://commento.io and Talkyard https://www.talkyard.io/blog-comments (I'm developing it).
Can you configure the dev branch to show up on the landing page of github?
Settings -> Branches -> Default branch
We have our develop branch set up, which tracks all the contributions as you would expect. We treat our master branch as a latest release branch.
From prior experience, Vox outlets are extremely stringent about what they permit in their comments: specifically, only people who agree or can extend the arguments of a given article further are preserved; or quibbles about minor details. No true discussion is allowed outside the article's bubble.
The upside is there, if you know up-front its the basis of the engagement. If its applied retrospectively it sucks, for a number of people.
But curation is not always bad. "it depends"
Vox itself removed comments after a time, probably because this bubble could not reasonably be preserved in response to the majority of their articles.
The Verge is somewhat more open, depending on the topic and whether it is technical or political in nature.
Polygon's Ben Kuchera is a caustically aggressive moderator, and is the primary one I saw behaving in this manner, in addition to antagonizing commenters who disagreed with an article's premise; doubly so if he himself wrote the article.
With this recent product that is also closely integrated with Vox properties, I'd say this is a deeply concerning move how closely the two are becoming entangled.
These Vox-alikes seem to fancy themselves arbiters of meaning with regard to “journalism” and “democracy”. Anything which disagrees with their personal definitions must be “alt-right”...
(If you happen to know)
Here's a blog post about the change: https://blog.mozilla.org/blog/2019/01/22/the-coral-project-i...
It writes: "After 3.5 years at Mozilla, the time is right for Coral software to move further into the journalism space, and grow with the support of an organization grounded in that industry", and "... The Coral Project will receive the backing of a large company ...". — However, Vox could back the project, also if it was still led my Mozilla, right. Still unclear to me why Mozilla didn't want to continue leading the project? Maybe Talk was too different from other things Mozilla does, and maybe it cost money and didn't bring any revenue?
I think it's dangerous to set hard requirements for such labels, specially ones like "journalist". It's not a given that journalism is considered worth protecting and, while in most western countries there's a sort of public agreement that "journalists" deserve protection, this often seems to be associated more to the word "journalist" than the act of journalism itself, and taking that label away from someone may also leave them defenseless, as they no longer "deserve" the protection of a "journalist".
It's more contentious than that. Journalists supposedly deal with "facts", but there's often considerable disagreement over facts or their interpretation. No matter your political persuasion, I'm sure there are "journalists" making a living on this, as per your definition, but who operate on the opposite end of the political spectrum from you and from whom you'd want very badly to strip that title of legitimacy.
There has been a lot of writing and thinking and discussion around what makes a journalist, and different styles of journalism, for some time now.
Ultimately the challenge is differing views of reality - disagreement about truth. I think that’s the root issue.
I've wondered why most publishers of comments don't just give up and say "comment on Twitter." Not that I would like that, but still ...
Here an example:
https://nicolas-hoizey.com/2017/07/so-long-disqus-hello-webm...
The missing part is the creation of the initial tweet in your idea, but that should be easy to automate.
I think that wouldn't work because of copyright and Terms of Use. Probably somewhere in Twitter's ToU there's something that says you cannot copy Twitter discussions and add to your own website. — If you study Twitter's ToU, probably the people who post to Twitter, never gave you permission to include their tweets on your blog.
I like the idea, and would want to do sth simiar, but just linking to tweets, instead of "downloading" them to the blog. And the blog author could write a summary of what the Twitter discussion is about (what sub topic related to the blog post, the Twitter discussion discusses), next to the link.
I'd also like to automatically find and link to discussions at HN and Reddit and Mastodon (in a commenting system I'm developing, seem my profile)
Re: explicit algorithm violation notices: Good, but it does not sound like there's an appropriate appeals process in place, something required even with algorithms in place. Also important to have said algorithm be tuneable, or even more preferably, pluggable.
And possibly make every judgment public and transparent like the courts, but redact any content that is illegal and/or involves doxxing and other privacy violations. Whenever humans interact, we need something like a justice system/courts.
My experience in the past has been that comments are not taken seriously by the publisher, seemingly just added to tick the "social" checkbox. But to make a commitment to the dialogue an article can generate seems like a positive development for journalism.
And some of it is kinda trash, but they'd still be better replacements for Bret Stephen's 100th "Democrats should listen to me, a person who will never vote for a Dem, on what to do" article
Is the reason possibly AGPL Javascript? Maybe then server side code, could be AGPL, and Javascript instead e.g. MIT?
However don't one typically place a blog commenting system in an iframe. Then personally I think you wouldn't have to release the website that embeds the blog comments iframes, under AGPL? And a discussion forum, one places on a sub domain, on a separate server? Also separate from your main website?
Background: I'm developing AGPL discussion software that you can also embed in an iframe, for blog comments. https://www.talkyard.io/blog-comments. What are your thoughts about using sth like that, if the code that creates the iframe is MIT or Apache2, and the code that runs inside the iframe is AGPL?
From a commercial perspective, I think you would be better off with the AGPL as well (perhaps dual licensing). I don't think you'd be able to sell such software on a per-license basis anyway (or, at least you'd not make much money at it as the audience is pretty limited). I think you'd be much better off positioning yourself as an expert in the area and selling contract services for other (perhaps proprietary) systems. Cheaper to hire you than to implement it themselves while tip-toeing around your implementation which they can't infringe.