Social media is broken – a new report offers ways to fix it
mitsloan.mit.edu
mitsloan.mit.edu
For a new social media to be "fixed", the company behind it needs to have the incentives and business model to steer towards more healthy behavior. None of these cover that.
We've had plenty of social media startups try to fix social, like Path, but ultimately they've failed bc a "fixed social media model" almost seems antithetical to "a good business."
I'm all for Pinterest and I think it's relatively healthier than the rest, but they barely have a functional business...
Signal would be a good example in messenging space.
dApps still on it's infancy, but it might come... Not sure if anyone will use it though, like the rest of the blockchains, lol
That's what the entire internet was like in its early days.
Ads killed the web
On an actually decentralized platform the radicals would still be there, but so would many other voices. Where users can't be banned for "wrongthink", communities must convince their audience with arguments or fear losing supporters. The chilling effect would be gone. A truly decentralized site would have plenty of communities the individual user doesn't agree with - just like in real life. And just like in real life users can deal with it by staying away.
This makes no sense and is completely wrong.
Can you give some examples?
Were you ever on Usenet?
Usenet was rad!
Usenet failed to scale past the first million or so users. Present social media platforms number in the 100s of millions to billions.
I've argued for several other reasons in addition for which Usenet died:
https://old.reddit.com/r/dredmorbius/comments/3c3xyu/why_use...
For the longest time I paid for MeWe, hoping they'd introduce pseudonyms and become a non Google Google+.
I would actually have paid for Google+ too if they had kept it around.
I'd like to discuss mental health issues as well but personally I don't feel like posting to ADHD or anything when my family or my boss or my future boss can trivially easy (two clicks, no mental gymnastics) go from my post there to my professional profile.
Besides, the term "far-right" is so utterly abused now that it doesn't mean anything except "people I don't like".
And, hey, would you have been happier if I'd called them Nazis?
I only paid to support a good cause and in the hope that it would become a replacement for Google+.
After a year or two where they never responded to feedback I just gave up.
I don't need another place to worry about leaking my identity connected to deeply personal issues.
Because big media got hooked on advertising revenue and news became subsidised and we got news/information cheaply, we'd been collectively taught that fact generation and dispersion costs were neglible (although they're not, work and energy is required and this was offset by the ad. revenue).
Move this to the digital space where information flows at close to light speed and you can accurately measure all sorts of marketing stuff - it just goes faster to the point of absurdity. Whatever value we can create via connectivity gets trashed by being subverted or obverted by corporate interests.
I'm now pissed off because I pay youtube $x per month not to be interrupted by ads, and now the content creators have their own ads in their content.
When you scale out, it becomes a sad picture of humanity that is excitedly self-reinforcing monetisation of things via interruption of engagement/continuity.
Actual value is evapourating.
No wonder we're all getting ADHD.
The spaces that are mostly simple and left alone are where interesting things can still continually occur. That's precisely why HN is actually still going strong.
When a tribe gets to large, because of Metcalfe's law the tribe loses the ability to discuss things to a level of nuance that is required, and that causes fractures that lead to a tribal split.
Because all of the internet communities have no natural way to split, the only thing left is for them to become toxic and eventually die.
Systems attempt to keep the tribe going longer by attempting to keep track of metrics, and this works for a while because the tribe had built up trust, but the bigger the tribe gets(with new members never trusting, forming Eternal September), the more the metrics reign, Goodhart's law states the more inevitable it is that it collapses.
> ... there seems to be only one cause behind all forms of social misery: bigness. Oversimplified as this may seem, we shall find the idea more easily acceptable if we consider that bigness, or oversize, is really much more than just a social problem. It appears to be the one and only problem permeating all creation. Whenever something is wrong, something is too big. ... And if the body of a people becomes diseased with the fever of aggression, brutality, collectivism, or massive idiocy, it is not because it has fallen victim to bad leadership or mental derangement. It is because human beings, so charming as individuals or in small aggregations, have been welded into overconcentrated social units. - Leopold Kohr, 1957
Cafés and restaurants handle this well. I think it is because they “commoditize their complement”[1], i.e. you’re allowed to talk while you consume. (Not too loud, though.)
A café, in most places, is a place for hanging out that makes money on the assumption that people will like refreshments, snacks or even small meals while doing so. They're typically not going to ask you to leave just because you haven't gotten anything in an hour or two (unless they're full). The exception appears to be some places where coffeehouses think they're a theme park ride.
How do you beat free / cheap food?
This is the opposite of what I learned in Econ 101. People voluntarily transact because both parties stand to benefit - otherwise one party would choose to sit out.
Money is meant to represent equivalence of value but it doesn't do this very precisely or well - which is why using it to 'measure' things causes so many problems.
I get where you're coming from, but that's Econ 101 idealism where you assume both parties are fully knowledgable about everything regarding the transaction - the real world is not like this. People also naively transact simply because they've been taught to do so.
I think regulation is the answer that aligns with the business model. All businesses are forced to align in the same way, and no company can benefit from breaking the rules.
If you fine companies for every day foreign bot accounts use the platform to spread misinformation, they'll staff up on countermeasures.
If you pass laws against amplifying fake news and spreading negative sentiment, there will be machine learning investments like we've never seen before. Comments that have low confidence scores won't be shared broadly.
If you regulate these companies like common carriers, they'll respect free speech on all sides of the aisle as long as it isn't law breaking.
Different jurisdictions are going to have laws so divergent that a single platform won't be allowed to operate globally anymore.
So we'll have to have many different providers that can all talk to each other.
Maybe this will be mandated by fiat. Or maybe it will just evolve naturally in response to incentives.
There is no incentive to fix that because it isn't broken for the likes of Facebook. Of course for the rest of us there is a huge incentive. But it raises the question of how. How is it going to work, and how are you going to convince people to use it. The first part has lots of answers in the form of social networks without a lot of users. So, the latter part is the problem.
As for authenticity and integrity; some notion of using cryptographic signatures could work. It's not particularly hard. News especially should only be coming from authentic sources. It's such a low tech solution to just sign your work and stake your ruputation. But somehow that's not a thing. Instead, our social media feeds are full of crap from dubious sources. Because click bait works and sells clicks.
That points to a solution. The likes of Facebook should start authenticating sources of information and start accounting for reputability. They are obviously incentivized to instead serve you clickbait. So the solution is to incentivize them otherwise. Hold them accountable. They help spread misinformation and profit from it. That could have consequences. When it does, Facebook will adapt.
If there was a combination of facebook, change.org and https://voteflux.org, allowing people to directly vote about issues they care, trade votes, propose laws, create local communities with custom laws, and in general control the way government spends taxes, there will be enough money saved to allow the companies providing this service to be richer than facebook without using any shady tricks.
————————————————————————————————————————
So, I'm totally okay with this comment gettin' downvoted, but can ya at least like … I dunno … add to the conversation maybe?
Perhaps a small comment on why his idea isn't pure genius? Or why I shouldn't find it amusing? Or how about why replacing politicians with a better system could somehow be a bad thing? Downvote me if you like, but downvotes without context don't help me (or anyone else) understand anything you might be trying to get across.
————————————————————————————————————————
Just for fun, I'll propose another idea; Let's start drafting people into political office.
"Congratulations {random citizen}! You've just been elected President!"
And just like a regular wartime military draft, refusal to do the job to the best of your ability = jail time.
Democracy doesn't work because the masses are smart and get to have their say. They work because democracies are an agreement that we can do anything in our power to change the government EXCEPT violence. Making the government more responsive to the population will result in worse government, not better.
I'm not advocating for totalitarianism. I'm just cautioning against thinking that it's the wisdom of the people that makes good democracy.
Now issues like abortion/immigration/drugs are used like carrot by politicians to herd voters one way or the other, and politician elected to as a result of support of one issue ends up voting on lots of other issues in a way that absolute majority of voters do not like.
Voteflux being a marketplace of votes, can allow people with different beliefs to find compromises with each other, will allow politicians to understand what people actually want instead of guessing based on number of angry letters/tweets, and will allow people to organize in a way to have different laws in different places allowing society to experiment with different things instead of fighting life and death battle each election.
So it would be a turbulent time, but no violent revolution.
I can't think of Pinterest as more than pollution of Google image search. They may be good, but the first impression is so bad that I don't want to go further.
There is a difference between providing infrastructure for social connection and farming/harvestings/leveraging the information gathered from what is constructed on that infrastructure.
The current situation is akin to if AT&T had listened in on all phone conversations (assuming no technical barriers) and sold ads based on what you said to friends and family. Before you made a call you would need to listen to a personalised ad on your reciever.
Before facebook or myspace or twitter there were blogs/personal websites, forums, email and instant messaging. All these options worked for expressing one's self online and none of them were under a centralised, for-profit entity.
Wikipedia is a good example of a decentralised social network although it appears to be increasingly infected with the reddit/twitter virus.
These are the same policies Wikipedia has had for over a decade, with only relatively minor changes. Are you saying they're being interpreted differently? Could you give an example?
Anyone suggesting Bellingcat is more reliable than Wikileaks is regurgitating establishment talking points.
FWIW I don’t think it’s an MI6 backed site
at least US intelligence agencies used to hide behind private news sources until they started hiring the intelligence people outright.
They're a state backed propaganda machine pushing the state's narrative, but resisting any other narrative when the current opposition in power because they've been entirely staffed by activist journalists.
The BBC works directly with the UK Foreign Office to push propaganda to destabilise other countries: https://thegrayzone.com/2021/02/20/reuters-bbc-uk-foreign-of...
> with only relatively minor changes
Relatively minor changes have huge effect on the content of articles. The linked article also addresses how Wikipedia has banned conservative sources from Wikipedia including Fox News, the Daily Mail, and the New York Post. "In short, and with few exceptions, only globalist, progressive mainstream sources—and sources friendly to globalist progressivism—are permitted."
Every claim in a Wikipedia article must be accompanied with a source. Claims that are only covered by conservative media and not covered at all by mainstream (liberal) media, cannot be referenced as a source in a Wikipedia article. This leads to conservative viewpoints being removed from articles. Which directly causes articles to become biased towards mainstream viewpoints.
That checks out.
It’s clear based on user behavior that users like the experience that centralized, for-profit social media networks provide. That experience invariably is very expensive to deliver… Dramatically more expensive than serving Wikipedia, for example, and the costs grow faster as the number of users increase.
So, in order to deliver the user experience that real people have demonstrated their preference for, one of the key inputs is exceptionally large (and increasing) amounts of money.
In order to be sustainable, that money has to come from somewhere. Targeted ads aren’t the only option, but no viable competitor has been discovered.
The reason why targeted ads are so productive here is because the value of the network grows non-linearly (up to a point) with each additional user. If you charge an entrance fee to each user, your network will be smaller and therefore outrageously less valuable than any network that does not charge an entrance fee.
Therefore, so long as any social network is willing to make their network free to enter, effectively no network with a comparable service can charge for usage.
Importantly, not only will the free social network be more valuable than the paid social network in relative terms, it will also be more valuable in absolute terms – both in terms of value received per user and number of users receiving value. So, it may be possible to regulate free social networks out of existence, but doing so would necessarily destroy a lot of user value in the process.
Perhaps there is some type of patronage or donation-based model that could both enable the free entry price and avoid the pitfalls of the ads-based business model. However, an organization like this is unlikely to attract the technical talent necessary to build competitive products.
Of course, like both of us mentioned, other models and competitive ecosystems exist. Centralized social networks have not killed email or instant messaging or blogs or websites. Those things are, by every measure, actually way bigger than they ever have been. So in a sense, the thing you’re hoping for is already here, it’s just not the dominant modality chosen by other people.
There are fair questions to ask about the power that ownership and control of a centralized social network imbues. Though, I think those are orthogonal concerns to the question of funding mechanisms.
In sum, I think the world you’ve expressed desire for does, indeed, already exist. It happens to co-exist with the world that you have expressed distaste for. While I am empathetic to your distaste for social media’s influence and business model, I think it’s important to recognize both why it exists and the ways in which it’s existence is not strictly zero sum.
I am confident if a free simple social network had caught on among college users about the time wikipedia started then facebook or myspace never would have been commercially viable.
Almost all users prefer no ads and privacy to ads and no privacy.
I disagree that that experience needs to be expensive to deliver. Imagine a decentralised social media network hosted on its users compute devices, kind of like email used to be. Aside from large streaming videos not that much processing power or storage is required per user.
Further I do not know if enough people would be willing to pay for access to a social media site for it to ever reach millions of users. Even at 1 USD a month per person people wouldn't join in unless millions were already there to make it worthwhile.
I don't think this is a given, and it's not that expensive to deliver either - the problem is more with the de facto monopoly of network position. But there are lags on all these kinds of activities, and user behaviour today, is conditioned on user experience yesterday, it may not predict well user behaviour in the future. Social media is still new, the experience of having it delivered to a mobile phone barely 10 years out of the box for most of the world.
Consider a marketing team that predicted the future of the book by looking only at input from the reading habits of a kindergarten, and ignoring any correlation with how those would change as the kids got older. Note that none of this has anything to do with payment per se, and neither does today's social "advertising enabled" media.
You make the comparison to AT&T. They charge a lot of money for the service that they provide, and as you mention, they don't do all the shady things that social platforms do.
Most other great tech products are provided by businesses. Apple is the prime example of a company that is both (a) massively successful and (b) not hostile to their users' privacy.
If you're listing those as Facebook alternatives, you misunderstand the key ingredient that made Facebook get adopted by billions of non-technical mainstream users: the real name identities.
The other communications platforms of dial-up BBSs, USENET, Geocities, personal Wordpress, vBulletin/phpBB forums, AOL AIM ... do not have strong real-name identities for non-tech people to find each other. Anonymous/pseudonymous online handles and cryptic login names have too much friction of discovery and are not scalable to a billion users. (Previous comment about the real Rolodex acting as database primary key: https://news.ycombinator.com/item?id=15294086)
Facebook (possibly accidentally) bootstrapped the psychology of users creating logins with their real name in 2004 because Harvard students want to find each other on the service. In turn, as other schools were added on, other students were motivated to use their real names to interact with friends in other schools. The social graph of real names becomes so viable that old high school alumni that were "lost" can now find each other again and grandparents can easily now keep track of grandchildren's baby photos. The fake names of USENET and other forums prevent that from happening.
Emails don't work for social network growth because you can't derive an email address from a real name you know. Someone has to tell you the exact spelling of their email address. With Facebook, you can just find them by their real name (or their real phone number) and send a "friend request".
>All these options worked for expressing one's self online
The secret sauce to Facebook's dominance is its accumulation of real identities and not because it gives Facebook users a way to "express themselves". E.g. A lot of grandparents following grandchildren photos don't post anything.
Consider direction of cause & effect:
- easier evolution of platform: a bunch of users login with real names to follow/poke each other --> easy to build on that network into groups for discussion, shared calendar events, share news articles, etc
- harder to grow platform: start with discussion forum software with fake names --> impossible to build a giant sticky social network graph from that base
Facebook & Linkedin built on real names. WhatsApp growth built on real phone numbers.
Platforms built on real identities will naturally centralize because nobody has come up with a viable decentralized alternative for a universal identity database. Ideas like web-of-trust, blockchain, keybase.io, etc don't accomplish the same thing as Facebook's real_id database.
[To downvoters, if you have a better explanation of how Facebook got mass adoption by non-tech people in ways that didn't happen with USENET and vBulletin web discussion forums, please reply with your thoughts.]
Finding people you know on Facebook is only one aspect of their success and it's probably the most heavily network dependant one. The other thing that drove adoption was the relative simplicity of publishing content. Your Facebook profile was/is essentially a dumbed down webpage where you can post text, links and photos. You also get free hosting, free "email" in the form of messaging, and a bag of other features that you would normally have to set up yourself and pay for. They did a good enough job of providing all of the killer features of the internet for free to average, non-technical users.
Because it's hard to get people to enter that information. In contrast, Harvard students in 2004 were highly motivated and that started everything off.
From 2005 to 2008, the killer feature of Facebook was entering your real information when signing up and then Facebook showing all the real people in life that were connected to you. E.g... old high school friends finding each other again. Likewise, signing up for Facebook means you also wanted to "be found" by others. It was the first mainstream service to successfully accomplish this. Myspace didn't do it. Friendster tried but their webpage performance was too slow.
>The other thing that drove adoption was the relative simplicity of publishing content.
Most Facebook users do not "publish content". Most just follow their friends or read links passed on by friends (or the algorithm). [EDIT for clarity: I mean "publish content" the way that gp I replied to cited "blogs/personal websites"]
By publish content I mean post links, photos, comments, rants, etc. Do people no longer do that on Facebook?
It's not like facebook required one in the beginning either. You could signup with joe s and many of these profiles still exist.
The secret was the rollout (the school network connected to your edu email), and other tricks like getting you to give your login/password so hotmail contacts could be downloaded.
The journey to the top has a lot of little wins. Real names were a step added much later and connected with their ad network strategy.
I understand but my comment wasn't trying to explain Facebook competitive market value by reducing it to only to real names. To beat Friendster, beat Google+, and get to its successful IPO, one can point to many things ... e.g. the addictive NewsFeed rollout in September 2006... and free unlimited disk space for photos, photo tagging of friends, etc.
My comment to gp's quote was how Facebook as a social network works very differently from USENET/Geocities/vBulletin/AOL_AIM. One can't point to those older communications alternatives and say they work "just as well". On those platforms, there's a fundamental difference that prevents people from even finding each other.
When they opened the floodgates it was a chance for normal people to associate with fancy harvard people and show how they were too cool for MySpace now.
You would be hard pressed to duplicate that kind of buzz anymore.
I run a Mastodon instance for myself and some friends. It's got about 250 accounts total. The hosting costs are about evenly split between myself and friends willing to throw me some money via Patreon. It is not making me money and I do not want it to make me money because then it would probably require a lot of my time to deal with moderating it. I don't want to have to do that or hire anyone to do it.
I run it to have a space that's free from the bullshit corporate social media does in the name of "increasing engagement" to make more profit off of ad views. I like that I can visit my instance, catch up on the local timeline and the stuff I'm personally following, and then be done with it for most of the day, versus opening up a corporate site that's had tons of very smart people sink a lot of time and effort into figuring out how to keep me there for hours on end, whether or not I'm better off for it.
The value is in the network infrastructure, not the so-called "tech companies" operating on it as middlemen, spying on internet users as a so-called "business".
We as users already pay for that infrastructure. We pay for an internet connection. The network infrastructure is already monetised.
That network infrastructure is worth something even if every single tech company ceased to exist. Would you stop paying for internet access just because some website went offline.
The large websites many people are using are 100% user-generated content, much of it being in the form of human communication. That so-called content is not going to disappear if tech companies stop spying for money. (We often hear of tech company users losing their content when the companies discontinue so-called "service". Not surprising since the companies serve advertisers, not users.) The tech companies are not the content creators. Generally, users do not pay tech companies for service and tech companies do not pay users for content. Tech companies are middlemen.
"Tech" companies want you to believe "This is how it must be. There is no other way." They want you to believe they are "the internet". But the menu is not the meal. The map is not the territory. The middleman has no inherent value. He must leech to survive.
People online often comment that "Facebook" (not to mention many other tech company websites) could be written, for free, in a weekend, and what makes Facebook so popular is not the people who write PHP for Zuckerberg, it is the network effects, the fortuitous timing and luck that Facebook has enjoyed to become so large.
What has been done to date can be replicated by others, but Big Tech wants people to accept another "default assumption": massive scale, all under single ownership, is a requirement.
AT&T was broken up and we did not lose telephone service. Around the world, customers of one telephone company are able to call customers of other telephone companies.
Local governments like cities / towns should provide federated apps like Matrix / Activity Pub as a service.
Your tax money should pay for the maintenance of the servers.
Besides CPUs that get faster and more efficient every year, phones that get smaller and last longer, global interconnected networks, cars that don't burn fossil fuels, ...
I ran my previous startup (Twitpic) completely on ads and it wasn't an enjoyable experience from the business side. It also does not (usually) align with the user's best interest. Ad business models requires attention to feed it, which in turn requires social media companies to build features to get as many eye balls for as long as they can on their app.
I also don't think paid only is the answer either for most cases. Some are ok with seeing ads in exchange to not have to pay to use. I'm curious to see if there is a balance that can be struck with services offering both options.
Brave's new search engine[2] is another example saying they will offer a paid option in the future. I used their beta and it's solid. I'd be willing to pay for it to be free of Google and help sustain it.
[1] https://glue.im/glue/glue-a-social-network-that-respects-you...
Glue looks cool. How does this compare to something like Mastodon?
I'm working on something related, called Circles [1]. It builds on Matrix for decentralization and E2E encryption. We're also in beta, hoping to launch later this month.
Circles looks cool as decentralized and E2E are fascinating to me. Glue doesn't have a whole lot in common with Mastadon, but similar microblogging features I believe.
Hit me up sometime, would like to hear more about your project. - noah@ark.fm
Yeah, like the man said, the greatest minds of our generation have been focused on getting people to look at ads. They're scary good at it.
1. The platform/provider/server needs to work for the users, not for some shady third party like the advertisers. The most obvious way to do this is to have the users pay for the resources that they consume. There are alternatives, too -- users could band together to form cooperatives, or some benevolent foundation could kick in a lot of funding. (This last one is the current model for Signal; I guess we'll see how well it lasts...)
2. Key to making #1 work is that all the content needs to be end-to-end encrypted. Otherwise the platform will be too tempted to also start doing targeted advertising, in addition to charging for access.
Neither of these is all that difficult. We have all the technology right now. Mostly we just need to get people together and do it.
Are they still going to pay when most people leave the service because they won't pay?
What regular person would rather pay then see ads? If you gave people tvs they would put up with the ads. If you give people free dialup internet they would put a banner across their browser.
Paying is a non-starter.
End to end encryped what? Posts/videos? Who cares..
Initially? Parents of young children, who want a safe and easy way to share photos and videos with their friends and family.
After that, people who enjoy having a nicer experience than the garbage tier platforms we have today.
> What regular person would rather pay than see ads?
Everyone who pays (or has paid) for Netflix instead of watching network TV. Before that, everyone who paid for cable when it was ad-free.
> Paying is a non-starter
That’s what they said about music after Napster, too. How’s Spotify doing these days?
I don't care how many people have netflix, spotify or cable when I purchase nor does it affect my experience. I watch the same movies if you have it or not.
If they charged for youtube less people will signup, less people will make content, more people will leave more content producers will leave. The values goes down.
Parents may buy space private space for photos/videos. Dropbox, icloud, google drive. But that isn't necessarily social media. Twitter, facebook or instagram would not survive on parents buying children accounts.
"How’s Spotify doing these days"
Doesn't spotify have ads? You could buy a subscription.. you also could buy a subscription to youtube.
https://mitsloan.mit.edu/ideas-made-to-matter/case-new-socia...
Or putting it the other way around: The only way to make money with social media is by compromising your users' mental health.
They made almost $2 billion in the last 12 months. How is that not thriving?
No printing press is obligated to mass-print a book you write. Similarly, you cannot go to a corporation's campus and distribute stuff unless you have their permission — because it is their private property. Just because private parties are refusing to give you a platform does not mean they are totalitarians.
>> Totalitarians will always use an example of where speech somehow created harm to exhibit the need for control, the truth is that freedom does cause harm, and will always cause harm, because you cannot eliminate harm completely no matter the system.
This is a strawman in the shape of an Argument from Futility. It's no different than saying "why have safety systems in cars when you cannot eliminate car accident deaths completely?" It's a strawman because nobody is aiming to completely eliminate harm from online speech, just prevent it in egregious situations where it can be prevented. It's an argument from futility because harm reduction is still a valid goal even if complete elimination of harm is not possible or feasible.
This is off-topic, I feel. The article illustrates that there is a push to stop "printing press" from printing "your book" in cases when it's perfectly happy to do that. In the name of fighting misinformation, of course.
Anyway, it doesn’t matter, i can finally say I’ve found my people now, @thebirdsarentreal :-) (its a satirical take on belief systems that vehemently denies it’s satirical which is just chef’s kiss)
Satire and irony used to be effective. In last 20-years cynical distance and ironic posturing have become so prevalent that it is no no longer considered subversive. Citizens can consume outrage passively through various satirical media products, displacing outrage and abstaining from more active forms of resistance.
People consume satirical content, and think they are above it all, while just participating in cynical ha-ha.
Fast forward 10+ years and people actually believe it. Someone recently launched themselves in a homemade rocket to prove with their own eyes the flatness of the earth (and died for their efforts).
I love birdsarentreal, but part of me isn't looking forward to finding out what happens to it as more people come across it.
> People [...] sense instinctively that the [...] truths [...] don’t go together, that [the promoter is] not being logical, that [they're] not speaking in good faith. But they’ve been told that God is mysterious, unfathomable, so to them incoherence is the closest thing to God.
This tangles up with the appeal of being an "initiate", "on the inside", and knowing the secret truth, and the combination is powerful.
All that is needed is honest discourse and a good educational system.
But what twaddlevision did to US society in the pre-WWW period, social memia is now doing to subsequent generations.
The circle is unbroken. ^_^
One is the freedom of speech, the other is freedom of the press, we as humans have the right to both. Yet never before in history we have had access to the press power of the social media revolution. For a time, the gatekeepers took a hands-off approach, but that time is over.
End Internet censorship now. We need platforms where the battle of ideas can be fought and won on a level playing field without the creeping hand of censorship in the name of "combating disinformation" getting in the way.
Fact check all you want, may you win the battle of ideas. But talk of censorship, even from MIT, stinks of totalitarianism.
Because that's the problem with the absolutist position against "internet censorship." It denies, ever, the possibility of a harmful kind of speech which should in fact be acted against. People can list lots of examples. (Doxxing. Calls for violence against individuals. Child pornography. Dangerous medical misinformation. Copyrighted information.) Are we really okay with a billion people getting all of these things -- rather than one person being "censored"?
And the other problem is we're not talking about a government with a constitution... We're talking about Instagram and Pinterest (and other social media companies). I've heard it said that they very conveniently claimed "We're adopting the same absolutist free speech principles of a country" mostly as a ruse to keep from having to invest in any kind of monitoring of their services.
Lies don't end lives, people do. You can't say that a lie "caused" someone to do something. It may have influenced them, but they are ultimately responsible for their actions. More generally, I don't think we should be in the business of policing second or third order effects. Responsibility ultimately ends with the perpetrator of an action. If someone lies to one (or many) people, and some of those people go on to murder: punish the murderers, move on with life.
> Doxxing
Not sure what's wrong with this TBH - publishing public information shouldn't be a crime. Some sites have anti-doxxing policies, some don't.
> Calls for violence against individuals
Direct threats of violence are already illegal. People that make them should be arrested.
> Child pornography
Already illegal - find and imprison the pornographers. We already do rather a good job at this.
> Dangerous medical misinformation
In addition to the obvious idea that people are responsible for their own worldviews, any process to distinguish "misinformation" from "information" requires an oracle that everyone can agree to trust. We have no such oracle.
> Copyrighted information.
Reproducing copyrighted material is already illegal and pretty well enforced.
> mostly as a ruse to keep from having to invest in any kind of monitoring of their services.
Why do you think it's OK to impose these costs onto businesses? Why should companies who effectively provide a digital bulletin board be responsible for policing its contents? Why not have... you know... the police be responsible for policing? Then you can clearly and directly send any complaints about their effectiveness to your local representatives, who will actually be empowered to do something about them.
But if that's the case, then at some point a society is also going to need to consider how it's going to also address the distributing of those things which are illegal. (If something can still really be disseminated to a billion people -- then what was the point of even making it illegal in the first place?)
For decades and decades all publishers have been responsible for the content they publish. (See libel laws, just for example.) So it just seems irresponsible to now concede the existence of vast unpatrolled online empires making billions of dollars while simultaneously creating dangerous (and illegal) situations which others will then need to police for them.
Good! Then we don't need any new laws or policy changes. We just need to empower law enforcement to do their jobs. As I've mentioned, they seem pretty empowered already.
Publishers indeed have been responsible for content. Sites like Instagram or Pinterest are hardly publishers however, they're more akin to the bulletin boards that used to be ubiquitous in public places (With Pinterest it's literally in the name). Anyone can post whatever they want without any editorial process. If someone posted a murder contract on such a bulletin board, would the board owner (say your local grocery store) be responsible? Hardly.
Vast unpatrolled online empires where people can say whatever they want sounds pretty good to me actually. It's freedom of speech in action. Anything actually illegal or really dangerous is policed pretty well (say compared to TOR, and even there criminals are not immune). Otherwise just leave people be. No one "creates a dangerous situation" just by saying something online. It takes more than that.
In a world where the Russians have a professional and very active disinformation organization, your "level playing field" looks more like a military conducting a massacre against a civilian population...
I suspect that this is a false choice, akin to “Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety”.
If we may observe history, we see that societies where one is not free to speak do not have more “true beliefs”, and indeed believe many things we think of as nuts.
The Pennsylvania Assembly wanted to raise money to defend the frontier during the French and Indian war by taxing land. The Governor kept vetoing this at the behest of the Penn family, which owned a lot of land that would be hit by any such tax, each time finding some objection that Franklin and the Assembly found made little sense.
Franklin wrote to the governor arguing that the Assembly should not have to give up its liberty to exercise taxing power over the Penn family lands in order to raise money for defense.
Trading liberty for safety is the crux of social contract theory, so the absolutist interpretation of the quote is clearly devoid of worth. We can instead focus on "essential" and "temporary," and then perhaps it has value but we must then argue about the meaning of those words.
... Is the freedom to shout falsehoods with Facebook's megaphone without Facebook weighing in "essential?" How "temporary" is the safety we gain from Facebook tossing fact-checks on posts?
At the same time, I think people enjoy being outraged. You get an actual physical energy burst, and you can think how superior you are to the misguided idiots you are talking about.
I don't think social media is tricking us into this behavior, it is just helping us do it better. It would be good if we could figure out how to discourage the behavior in the first place.
This leads to an inherent advantage for the producers of misinformation.
You approach worked say 50 years ago, when most of us had limited sources of information and so had time to actually see and understand the refutations of most attempts at misinformation.
Nowadays, we've almost all got more sources of information competing for our attention than we have time to deal with. It is much less likely we'll see the answers to the misinformation, or have time to deal with them. Furthermore, with what we see algorithmically determined to maximize engagement, we are probably going to be shown more misinformation from the same or related sources instead of the refutation.
I think people agree on a lot of the dynamics in this space and what needs to be done.
Lies, half truths, and everything in between are presented in the most riveting ways possible and our monkey brains just can't help but soak up every last bit of it. If a person believes everything on social media then __they__ should be fixed, not social media.
A great solution is for people to see social media for what it is: 99% garbage that gives you a quick dopamine hit.
People should be made aware that on social media you are likely:
1. Reading misinformation
2. Viewing content that's created with the sole purpose to manipulate you into a certain view (Often times by a bad actor)
3. Reading opinions by trusted people who've been compromised by the above two points
Just like a great solution to the climate emergency is to stop releasing greenhouse gases, a great solution to war is peace, and a great solution to starvation is to eat food.
We have large organizations with great solutions tackling climate change, starving children, etc....
I guess maybe like war, spreading this message would hurt pockets that are far too deep to be challenged.
I just think that people would be happier and social media would be a better place if the public were educated on what it truly is.
Just like journalism has become compromised, so has social media. The solution isn't to abolish both, instead we need to find ways to strengthen their integrity and make them independent of partisan corporate or political interests.
Free speech has never been as vulnerable as it is now. This is a new situation that absolutely requires for new solutions.
No, you're the one suggesting we abandon "Western ideals of information freedom", so you need to propose what ideals we should have instead, and what practical changes those new ideals will lead to.
"Citation needed". Who says it can't?
Even HN suffers from this - but far less than other places. Poor replies to popular comments will receive more views, there are few detailed conversations, just people sharing their view and leaving, opinions from often very knowledgeable people can get lost in hundreds/thousands of other comments.
We often talk about regulations in terms of volume, but the actual details matter.
Online groups only work when they have a uniform identity, an ingroup, a particular set of shared values. They can have other ideas or values, but it needs to be clear to them that they should only bring up that group's values. Communities can also be toxic and stupid, but they do keep cohesion when they're all aligned on a core theme.
Are enough people big enough suckers to allow this to happen under these false promises? We will find out. If you do support measures around controlling what people see and say, at least do the rest of us the courtesy and don't pretend that it will bring about a utopia of free exchange.
In general this reads like a sickening melange of half-true diagnoses (yes, of course lack of competition is a problem) and naive appeals to the nanny state that don't understand what they're inviting. To whit:
"Renée Diresta, research manager at the Stanford Internet Observatory, said policy should also differentiate between free speech and free reach. The right to free speech doesn’t extend to a right to have that speech amplified by algorithms."
Yes citizen, you can say whatever you want. You just have to say it in this sealed, soundproof room with no windows! This is almost literally the author saying "You have the right to speak, you just don't have the right for anyone to hear you", which is nonsensical. Free speech is meaningless if you can't actually reach anyone with it. The right to yell your deepest beliefs in the middle of the Sahara desert isn't free speech.
But of course, this ability to silence "misinformation" is fine, because it will only be used against the bad people! Surely it will never be used to silence people like the author, because they're good and right! Never mind that this is already happening (Youtube is famous for silencing certain LGBT and sex-positive figures, as well as any number of left-wing activists, for example).
Social media behavior and design is driven by the users in no small fashion.
The big companies merely amplify the dominant behaviors.
Remember the Instagram, Facebook and Twitter linear timelines???
Yeah. I thought you did.
The companies merely made it more attractive by amping up what’s most popular or what we like.
Social media cannot be fixed by big tech.
It takes users to fix it.
It’s just a mirror for our basest instincts.
1. stop the spread of fake new: This has already been happening, most of big offenders are gone (e.g Alex Jones) but what should be noted is that this can (and has) lead to backlash. The fake news ive seen always comes from tiny accounts that have unexpectedly viral content usually taken and reworded from banned sources (Alex Jones et al), I don't think cracking down on the most prolific offenders will necessarily be the fix. The actual report is way clearer and thorough on this then the article.
3. Lack of regulation for social media companies: I feel like if their is global regulation (which there should be) many countries will just ban it (e.g China, Turkey and Russia) and still lead to the same balkanization.
5. polarizing Algorithms: I don't think slowing online interactions will solve this. If someone wants to be racist they will be racist. I think this will just bring annoyance.
6. better social media business models: they say that they worry that "the best, fact-checked information is available only behind a paywall" but that is already the case!
I recommend people go to the actual report, the 25 solutions is on page 16
Keep the riff raff out and have users real identities linked to payments.
I’d rejoin Twitter if it cost $5-10 per year
I will never use a social media platform that requires me to link my IRL identity. It's DoA for me.
Count me out.
Twitter isn't Twitter without the wide range of people it attracts, for better or worse.
Social media requires that the people you are interested in are there for it to have any value. Not enough of my friends are going to be willing to join facebook for >0 usd, which means I wouldn't be willing to join it either.
Twitter can maybe be saved here, but they would a lot smaller.
Sadly 10 usd/year isn't enough to keep out the riff raff, in fact it would probably entice people to see it as an investment meaning we would get ever more crap from influencers, "thought" leaders, etc.
From what I remember from Facebook SEC filings, they make on the order of $50-100 per US user per year.
Here they are:
1) Hold platforms accountable for designs that amplify lies.
2) Focus on and shut down prolific disinformation networks.
3) Use content interventions to nudge people toward awareness of falsity and accuracy.
4) Use accuracy nudges to crowdsource falsity labels so that algorithms can be trained to automatically identify lies.
5) Give researchers access to more data, using technologies such as differential privacy and institutional mechanisms, like data safe harbors.
6) Offer platforms incentives, like legal immunity, to share more data with researchers.
7) Create independent panels to oversee research funding by platforms.
8) Require robust legislative remedies to anticompetitive platform practices.
9) Create collaborative global institutions where discussions about governance can take place.FTC or FCC—to vet compliance with guidelines.
10) Establish independent U.S. statutory bodies—not the FTC or FCC—to vet compliance with guidelines.
11) Offer platforms ‘earned immunity’ from civil liability, e.g. under Section 230, if they comply with content moderation regulations.
12) Break up big companies, starting with Facebook.
13) Legislate interoperability as well as social network and data portability.
14) Emulate EU proposals for stricter anti-competitive penalties to deter corporate misbehavior.
15) Create programmability with the major platforms to encourage innovation.
16) Diversify the engineering core to include social scientists, cognitive scientists, and people trained in different types of histories.
17) Create ethics certification programs and degrees for AI designers.
18) Embrace friction to reduce the automatic nature of information diffusion and interactions with AI.
19) Regulate platforms for product safety; hold designers and companies accountable for the products they design and deploy.
20) Develop new business models including subscription and freemium models.
21) Tie consequences of legal violations directly to corporate executives, not just their companies.
22) Strengthen regulations by adding taxes, such as programmatic media taxes, to deter algorithmic amplification.
23) Ensure that users consent to data use; protect data privacy.
24) Distinguish between speech and reach--the right to speech and the right to amplification of that speech.
25) Limit corporate lobbying; enforce stricter campaign finance rules.
> 2) Focus on and shut down prolific disinformation networks
> 4) Use accuracy nudges to crowdsource falsity labels so that algorithms can be trained to automatically identify lies.
> 21) Tie consequences of legal violations directly to corporate executives, not just their companies.
> 22) Strengthen regulations by adding taxes, such as programmatic media taxes, to deter algorithmic amplification.
> 23) Ensure that users consent to data use; protect data privacy.
> 24) Distinguish between speech and reach--the right to speech and the right to amplification of that speech.
> 25) Limit corporate lobbying; enforce stricter campaign finance rules.
Ordinary internet users hate it because it represents a serious loss of autonomy for them, and a degradation of service. It makes the internet less free than it used to be.
The wealthy and powerful (and those who love brown-nosing them) hate social media because it's STILL TOO FREE. There's way, way too much information there that threatens their vice-grip on the world, and it has to go. NOW.
We can be eloquent, humanitarian, calm and polite, but the reality is that we can also be cruel, vindictive little animals that seek out to ruin everyone's day. There is no fixing social media because our nature doesn't allow it, and you can fix the business aspect, but not how the users utilize it, unless you restrict them to the point we can't call it a social media platform anymore.
TL:DR Fixing human nature isn't going to happen. The more popular a platform is, the more positive and negative voices you'll have in the platform, and we all know that we linger and give more attention to the negative more so than the positive.
1. know who you talk to if you want to stay anonymous, use an anonymous platform. HOWEVER, that guy you entered a twitter shitstorm with? might as well be some 12 year old, broken english, russian troll. that chick who messaged you? 56 year old chinese gypsy.
2. mark content type sarcasm is (imo) very important. but in the wrong context, lack of context, it can be horribly confused.
3. exact timestamp I don't want to see a post is from "few months ago" (hidden, in a small text similar to the background color). I want posts to be color coded by age & type.
4. exact definitions my language isn't the same as your language. even if we both speak English, there are different meanings to different cultures, regions, etc. if something is ambiguous or unknown, I'd like to get a dictionary definition as reference.
5. I don't care about likes. or views. I want to see views/likes ratio. if a post has 1000 likes is it good? maybe, unless it has 1m impressions. or it's exceptional if it has 2000 impressions. that's a far better measure for quality
and there's more if I just keep thinking about it. probably a lot more
2: marking content as sarcasm completely undermines it's impact. 'sarcasm tags' are unutilized due to lack of purpose, not lack of visibility. how does facebook's 'react emoji' functionality not cover this usecase better?
3: you don't want the human-readable age, you want the exact absolute date, but also the age, which you want to be exposed as color..? absolute date is generally easily grabbed with js, but hidden to reduce clutter.
4: a product to put cross-culture communication up front would be an interesting prospect, but i've not seen anything like that, and i'd expect a full team, many revisions, and a patient and involved community to get functionality and ui approaching helpful...
5: you don't want the exact views and like, you want a human-readable measure for quality that you just came up with and can easily do in your head? (vs how you feel about predigested data re #3)