If your website's full of assholes, it's your fault (2011)
anildash.com
anildash.com
I was part of a now-dead "threaded message board", which had a somewhat cliquey view of what should/shouldn't be posted. In the evening when the moderators were away, the discussion became more interesting as it strayed off-topic.
Eventually the community wanted 'to fork' so I decided to build something myself. I'm not IT trained, so first version was PL/SQL on Oracle - all I knew. v2 was the product of me buying an O'Reilly PHP/MySQL book (I still have 'platypus' somewhere).
People came.
First few years were a mess. The more technically literate people used to hack me for fun - but then would confess and provide security tips.
As the mood took me I'd add features, and then if they became troublesome fixed/killed them.
The odd person would arrive and cause trouble, and I'd have fun working out how to make them stop. (correlate logins with emails, password, IPs, browser fingerprints etc). Then maybe experiment with hiding troll-posts from the other users, whilst displaying it to the troll, who thought they'd been ignored - etc etc. Or maybe auto-embed some very dodgy zero-sized images in their feed, if I know they were going through a company/university proxy.
After a while it just all got quiet and relatively happy, and it's been puttering along since 2003.
It's currently got over 9 Million posts, a load of marriages and children - and a few deaths with some 'best-of posts' for the departed.
I'm fully aware this isn't impressive as a potential unicorn, but I feel there's a bazillion other little sites, like mine, happily puttering along out there that are overlooked.
My little site is 'local' - I don't have to deal with the world turning up.
I don't need admins, I don't need a TOS, I don't need arbitration systems etc etc.
There's a good community of people who just get along. Rough edges were knocked off a long time ago.
We're a little niche ignored by the world - and we like it that way.
Issues with 'toxic' influx come when you're trying to drive people in/draw attention.
Mine? - Mainly what people have eaten for lunch...
..but if I had to come up with a proper answer it's that it's connected a pretty disparate bunch of people who once had a random common interest, over what happened to them in the next decade or so.
A shared diary?
Hopefully that's enough to work it out, and this thread has moved down the board to just the people looking at their replies.
Edit: I know it's gauche to complain about downvotes, so I'll explain my question further. I find his situation extremely interesting as I associate Oracle with the "IT-iest" of "IT" things.
Always loved computers, but unlike maybe most people here, for what I could do with them (not the beauty of their logic blah).
Did a Masters in Bioinformatics, that came with a few Oracle lectures... and then as this was in 2000 was enough to get a job.. and well I was given some problems and there was a shelf of books. shrugs
Pretty sure it wasn't - due to the piss-poor design of Y-combinator's forum.
"I don't like click" hides the thing you don't like.
Can't think of anything worse.
I genuinely think you should feel proud for making the world just a little bit better and more like it's ideal self
Every f'in part of it I built from scratch. It's me, for better or for worse.
More pragmatically, it helped shape what I like doing for my job (I'm currently a Product Owner - a job I think fits me)
He makes a really good point that there are lessons to be learned from, "disciplines like urban planning, zoning regulations, crowd control, effective and humane policing, and the simple practices it takes to stage an effective public event".
But I think what he's leaving out is hierarchy. In 2011, I think a "moderator class" and a "commenter class" was all you needed. But if you want people to treat each other well online, you need something more than that. There needs to be stakes. There needs to be value in building up your standing in the community, there needs to be something to be gained and lost in exchange for the community policing itself.
Right now, a community like Twitter has a sort of chaotic hierarchy. There is the average user, the influencer, the blue checkmark, and the Twitter employee/algorithm doing moderation. Who is answerable to who? How does one climb up the hierarchy? What value does a blue checkmark have in moderating an influencer? Can they? Are average users ultimately incentivized to be dicks in order to get the attention of influencers and blue checkmarks?
Maybe it's time for communities as big as Twitter to make some decisions about this and tame their userbase with more rigid and well-defined hierarchies.
This.
One thing that could be very effective is screenshotting genuine brand ad next to vile screed.
How long would it take for a 'bot to look like a white sheet wearing moron to YouTube, and then a little bit of sentiment analysis and you could automate this. Just publish the "daily worst brand fuck" on FB or YouTube.
(Alongside the "monitor tweet feed of politicians for death threats and pass to the cops" idea I have when I get the time I am really honestly going to ...)
This just happened recently on YouTube: there was a big uproar about sexual-ish comments being posted on videos of children, such as linking to timestamps with upskirt glimpses or similar. Advertisers got contacted about it, and some of them said they would stop doing any advertising on YouTube until it was addressed somehow.
In response, YouTube completely disabled all comments on videos featuring children, and deleted all existing comments. Now there are legitimate channels related to children that had completely reasonable comments (which they moderated) that no longer have any ability to have discussions on their videos. Their ability to interact with their viewers/communities was destroyed because YouTube decided to go scorched-earth to make sure they'd satisfy the advertisers.
..and??
Putting myself in YouTube's shoes, I've no idea how I could easily stamp out these links/comments.
What's maybe more interesting to consider is that there's been no blocks linking a URL to a particular time in a video - so it can still be done, just not from a site with a google domain in the browser-bar.
This sounds like a good mitigation step to me, considering that we're talking about child predators. The incentives are aligned properly here.
You sure about that? To me it seems like YouTube's incentives are to minimize the chances of advertiser irritation, while the incentives of the children would be to maximize their engagement while minimizing (the impact of) predation. Getting rid of children's channels entirely would be an acceptable outcome for YouTube, but I doubt the children in question would agree.
This is what Sleeping Giants have been doing for a while. They started with Breitbart... some of Milo's earlier articles were fantastic for this purpose.
To give an example, in its FOSTA-related policy updates, reddit also banned exchanges of heavily-regulated products like alcohol and firearms. There was no mandate to do so, as safe-harbor provisions are very much intact in case, for example, a reddit user sells a gallon of homebrewed beer to a 20 year old. Such changes might provide motivation for people who aren't at all horrible to move elsewhere.
> we have a way to prevent gangs of humans from acting like savage packs of animals. In fact, we’ve developed entire disciplines based around this goal over thousands of years.
We 'just' need to introduce institutes of online courts, judges, lawyers, and presumption of innocence (good thing that enforcement can be automated, so we do not need the execution squad). Also, you kinda need someone to write these policies, and approve them. Maybe an elected group of people - maybe even two groups, to balance the democracy/influence?
If you are not already doing it for your website, you are obviously an irresponsible moron.
I support this kind of stuff in theory but it is incredibly difficult to get to scenarios where this actually happens. You need to be massive scale for it to occur with any regularity.
In case of reddit, the problem is even more interesting. There are subreddits, so it is naturally segregated, and moderators and community are local. The communities can use downvotes to 'stone' an offender, and local elected moderators can enforce a ban. In case moderator or community mishandles the process, 'federal' admins can intervene. So we are pretty much at medieval level of justice there (much better than whatever constitutional dictatorship facebook and twitter have).
As I understand it, Riot Games had something similar implemented -- a jury of peers called the tribunal to render decisions in order to punish bad player behaviour (griefing, hate speech, etc) and it still didn't scale to the amount of incidents and abuse reports they were getting nor did it filter out frivolous reports. They ended up switching to automated systems as a first line of defense against obvious bad actors anyways.
Your feelings may have tempted you into making this off-topic post. The linked essay concerns comment section protocols, not personalities.
This isn't about the ad-hominem. The general point Dash is trying to make is fine. But it's easy to preach about not allowing assholes when you don't consider yourself to be one. The day the zeitgeist shifts considerably in the favor of assessing him as an asshole, I suspect Dash might sing a very different tune about site moderation policies.
What difference does it make who makes this argument, if the argument itself is sound? On the other hand, if the man is 50% saint, 50% angel, and 50% genius, what difference would that make, if the argument itself was terrible?
Comments suggesting that he shouldn't make this argument do not actually address the argument itself. In the best case, they are a distraction from the subject being discussed.
In the worst case, they are a form of tone policing, which is a tactic for silencing sound arguments by complaining that they are not being made by the right people, or in the right way. Example:
A man murders another man. The police torture him into a confession. He complains about the torture. Is he the last person that should be complaining about violence? But if not him, who? Must we wait for the perfect angel of an innocent victim of police brutality, before we act on it?
The general point Dash is trying to make about moderating communities is fine. But it's easy to preach about not allowing assholes when you don't consider yourself to be one. The day the zeitgeist shifts considerably in the favor of assessing him as an asshole, I suspect Dash might sing a very different tune about site moderation policies.
You're (possibly unwittingly) making another, an ad hominem circumstantial, by questioning his motives.
It's always best to just debate the argument itself and leave the author's character and motives and misconceptions out of it.
Otherwise we end up spending a lot of time discussing Mr. Dash, instead of discussing moderation of online communities,.
It isn't on topic, unless you want to say, "Since Dash is an asshole, that is proof that there exists at least one asshole on the internet, which supports part of the argument he makes."
The premise of the argument is that moderation is a moral imperative. Nothing about Mr. Dash's character supports or refutes that argument.
Simply state, "The problem with moderation is that it can be used as a hammer to ban dissent, &c. &c."
Put like that, there is no ad hominem, and it can be debated on its merits. By the way, the most amusing thing about that argument is that we would be debating it on a moderated platform, that we both use precisely because it is moderated.
If HN was unmoderated, it would cease to be valuable to the people currently using HN.
> that we both use precisely because it is moderated.
That's not true, I'd prefer if HN was unmoderated.
That is very interesting. You do not seem naïve, so I infer you you are well aware of the social dynamics at play in an unmoderated forum.
On such a site, if I say that attacking Mr. Dash for being an asshole itself is an ad hominem, somebody is going to say, "STFU, Nigger!"
A bunch of other people, whether because they're racists, or just enjoy the spectacle, will upvote that comment and downvote me to obscurity.
Somebody else will say, "Incorrect. He's not a nigger, he's a mulatto."
A third person will play along as the straight man. "What's the difference?"
The reply will come by swift rickshaw: "About twelve dollars in confederate scrip," and will get upvotes galore.
A follow-up will say, "Nigger, mulatto, whatever, they're all going to the ovens if they won't relocate out of the white settlements."
Communities that stop moderating quickly become toxic, because even the smallest amount of toxicity drives un-toxic people away, and attracts toxic people, and without anything to stop it, the trend accelerates, until the site is dumpster fire.
And it's not just the comments. People would get away with positing all sorts of political and social stuff, "because unmoderated." Many of the people who come here for the tech would leave, because this would no longer be tech-heavy.
The "truthiness" of the front page would slide rapidly as the site becomes an echo chamber, until it would be unrecognizable as Hacker News in anything except name.
I presume you know that, so I also presume that what you mean is, you prefer a site where we would never have this conversation, because I would no longer be here.
In fact, nobody would have this conversation, because everybody here would be united in their contempt of Mr. Dash, anyone like him, and anyone who doesn't immediately denounce anything he says on the basis of him saying it.
---
I am tempted to indulge in a little fallacious reasoning myself and say that what you want is not a "True Hacker News," because it would quickly become a completely different site with different people, talking about different things, in a completely different tone of voice.
But perhaps I am wrong, and what you want is all the same people, talking about the same things, but without anybody getting banned or any posts being killed.
If so... I don't think that is a stable state. Moderation is a local maxima, dumpster fire is another local maxima, and everything in between eventually gets pulled to one of the two stable states.
I think there's also room for inviting users to be moderators on a rotating basis. Anything is better than a centralized moderation team that can direct the tone and voice of a community in my opinion.
Thus, it is important to consider the motives of people who may end up enforcing them, because there's potentially a large difference between the law-as-intended and the law-as-enforced.
If the difference is large enough, it's worthwhile to examine how the personality of the enforcer may bias the outcome.
I think your argument might be valid in extremely rigorous fields like quantum mechanics - who cares who submitted the proof? But when we're talking about enforcing free speech, the person who is enforcing it has a big effect on the outcome.
(Even if someone write a computer program or book or movie or whatever but they are murderer and whatever, if the program/book/movie is good then it is good, anyways. And if it is no good, then it is no good, anyways.)
I interpret the comment as: "if the author can't follow his own guidance, or if the author's own compliance with this guidance is subjective, the guidance may not be given in good faith". The comment is not just saying asshole=wrong. It's saying that Anil Dash may be a hypocrite, and thus unable to propose good social policies.
The discussion (to me) seems more moral or social rather than technical. In cases of moral or social judgements, are you sure you completely reject ad-hominem points?
It seems to me that a fully Bayesian perspective requires considering ad-hominem attacks; if the source emitting the message is considered untrustworthy, should we discard that information?
[This argument] relies heavily on the idea that there's some universal category of "assholes" that websites should ban, and that if any website doesn't it's their fault and everyone should pressure them to fall in line. [However,] there is no such universal category that everyone agrees on.
That is simple, direct, and leaves him out of it. And I believe it is better without dragging his character and motives into it.
edit: Even if you have a policy that seems perfect to you, it will not be executed perfectly in practice when there are 1,000 different people attempting to enforce it as moderators.
Speaking of standards of “asshole” being different between communities, well, the website should have a set of rules/guidelines defining what “asshole” means, and those who aren’t happy are free to go to a different website or make their own.
If you have many different ones then you can see what is who like to post on what, I suppose.
Every site you type something into and press 'submit' has a complete choice as to what happens with your submission.
My take is that our current 'problems' are down to everybody wanting to record your interaction and take no responsibility for it.
That said, I think an ad-funded global discussion is impossible. You can't do good user moderation with that little money (~$0.10 per user per day total revenue, with a tiny fraction of that going to abuse prevention). I think Twitter should allow people to create semi-private spaces, zones for discussion where users can be much more firm about moderation. It should also make collaborative blocking and muting much more effective.
Everyone complains about Twitter but when alternatives are presented that gives them the power to do what they are asking for no one wants to try them.
One solution would be to not jam the entire world's conversations into a couple of gigantic, one-size-fits-all platforms.
You ask what solution would work for a platform like Twitter. Consider that the problem may simply be that Twitter exists at all.
You can see a hint of this in the final paragraph of this blog post: "Because if your website is full of assholes, it’s your fault. And if you have the power to fix it and don’t do something about it, you’re one of them." Either you purge the people he thinks are assholes, or you're one of the assholes who needs to be purged.
I.E. you could say, “my website is full of people, who I think are assholes, but I don’t know what to do!” and this could be a problem.
In addition, I find that “people who don’t care when they are told that their actions might be harming others” is a pretty easy way to define an internet asshole.
Which is ... everybody? People that argue for purging "assholes" will hurt those people's ability to communicate. Tell them that, they'll say "I don't care, those people aren't people, they are assholes".
If you zoom out and understand the overall goal of the community, then this kind of wordplay is easily neutralized by understanding: "What kind of relationships, conversations, and communities do we want to create?"
> "What kind of relationships, conversations, and communities do we want to create?"
You're absolutely right, it's just that the answer to this question is different for most people. There's no objective answer and in the end, you end up saying "these people here are more important than those people there". Which is fine of course, I just prefer that to be clear and not pretend like there's some absolute moral guidance we can turn to.
Yes, absolutely
> and in the end, you end up saying "these people here are more important than those people there". Which is fine of course, I just prefer that to be clear and not pretend like there's some absolute moral guidance we can turn to.
If you do not have an answer to the question, then you just use other defaults. It’s like how there’s no such thing as “unbiased journalism” - you either consciously choose your bias or unconsciously use someone else’s.
If you're a for-profit company, you have to pay them. At least minimum wage. See Hallissey v. America Online, Inc. They're employees.
I get annoyed reading the same smarmy-sounding "Please don't post unsubstantive comments" replies from dang all the time — but it's what makes this site work, when so many others don't.
Personally, I’d rather have the constant out-of-bounds reminders than not. I’ve never seen one of those warning messages on a comment that didn’t deserve it, I’ll take the ‘please play nice’ requests all day if it quells ugly flame bait, and that’s exactly what it seems to do. He’s giving people both notification that it’s out of bounds and advance warning that there could be consequences. Much worse is out there and easy to find. Do you have examples that are much better, on a comparably sized forum?
The thing is I haven't seen one of those in a long time. Now you can have entire threads where the only comments are "lol" or something like that.
Can't see why you were downvoted?
I also find "community" to not fit the bill, but that's another point.
https://www.metafilter.com/newuser.mefi
- Anyone can view the site and its content
- Very clearly putting their rules of the road up (link above) before you can 'join'
- Commenting/posting/membership is a $5, one time fee
- A one-week wait before you can post
- Paid moderation
And it works.
Instagram has a pretty positive feeling too.
Honestly, though, I think "major site" and "good community" are in conflict. Facebook's revenue is something like $0.12 per user-day. Twitter's is circa $0.07. It is at best extremely difficult to do effective policing of users with that kind of money to spend.
You can use ML, etc, to do the first cut. But real anti-abuse work basically requires humans. (Or not-yet-existant strong AI.) So my general take is that ad-supported social media is a dead end as far as real community goes. Doing it right takes either real money or dedicated volunteers. And probably both.
For example, had twitter not created a culture of basically brigading, they would be in a different place. But they at minimum allowed and probably incentivized that to become part of the twitter experience, and I don't think it can be recovered from.
As for facebook, if it took you until 2019 to say Alex Jones can't build his business on your platform, you haven't even started to think in any serious way about what behavior you want on fb. Well actually they have: mark wants dollars and doesn't give a shit about anything else.
You may find this interview with former Reddit pm Dan McComas fascinating: http://nymag.com/intelligencer/2018/04/dan-mccomas-reddit-pr...
If Twitter or Facebook had taken harm reduction seriously from the get-go, they wouldn't exist. Ad-supported social media only works at enormous scale, because of the network effect, because of the high fixed costs, and because of the low per-user revenue.
McComas is right that a grow-at-all-costs mentality is part of the problem. But that's a consequence of the financial situation, not the cause. If they had infinite money, they could have had growth and user/society safety. But they didn't, and they prioritized the things that let them get to the scale where the investors got the payoff that was promised. And where the people in charge got rich, of course.
I disagree, though obviously it's all guesses, and depends on your definition of harm reduction. I believe more strongly for facebook, but even for twitter, I think both hand pretty low-harm products inside them: flirting, hookups, dating, friends, jobs, photo sharing, etc, rather than chasing super high engagement via fake news. I don't include adtech as a harm in my case. I also think facebook had a ludicrously profitable Tencent payments business opportunity, had it not become such a toxic product and not spent all it's mindshare on adtech.
I further think both had a OneLogin opportunity: login with serious 2fa and probabilistic risk is hard, and selling a saas login plus social network import is (still?) a big opportunity.
I also think both would have been protected by the gab or voat-type cesspools: when you explicitly choose the people too vile to be on Twitter or reddit, that reinforces and magnifies itself rapidly.
And finally, FB busts ass for a $0.12/userday in the US: 45/year. I can think of a lot of businesses that are worth $4/month.
In other words, it didn't actually empower women. It just tolerated women venting about how very, very angry they were at men and, to some degree, taking it out on male members in AskMe who dared to ask questions about their love life. One particularly ugly incident involved a man asking for support resources for himself to cope with his wife having been raped. He was basically told he was an asshole for asking.
I wrote a long reply with actual suggestions which was deleted by the mods as "tangential" and "not answering the question." I posted it to my blog, where he soon found it and private messaged me and thanked me profusely and told me how traumatic it had been to basically be told he was an asshole for seeking support, a comment that got favorited at least 70 times.
There are no perfect communities and a good community experience is partly up to you to craft for yourself. Some Reddits seem to be decent places with a sense of community. I've been on email lists that worked for me for a time.
Hacker News had a genuine sense of community when I first joined. It seems to have less of that these days, partly a product of scale. It's grown a lot and it's hard to keep a sense of community at scale.
Even in IRL communities, large cities rely more on formality to help make up for the inability to know everyone.
The human mind seems able to have a sense of community with a group of up to 150 people. I had access to information from an internal study of one community that suggested that only 20 percent of members were 'active' participants, either regularly or occasionally. Another ten percent participated once or rarely and the rest lurked. This seems to fit with what I have observed elsewhere.
So that means that about 750 accounts gets you about 150 active participants. Anything beyond that and you see challenges in fostering a sense of community.
I've long speculated that HN's leader board ought to be 150 names instead of 100 to help foster some sense of community here, but it's a completely untested hypothesis and not anything I feel strongly enough about to actively campaign for.
You may be right about the value of such places, and Anil Dash's prediction that such places will be full of assholes may right, too.
To remove illegal content and conduct? Sure, but not to prevent assholery.