Tim Berners-Lee: We need social networks where bad things happen less
theguardian.com
theguardian.com
The problem with social networks, is that they're social.
Ephemerality being a big one. If there's no history, there's less of this cached aggression potential by raging over words said in the past. Ephemeral constructs focuses the conversation to the now since whatever current context you're in is fleeting.
Internal fragmentation (siloing) within a larger network being another. Silo might be a dirty word right now but I think combining ephemeral spaces with more siloing can help reduce the "bad things".
For all of the things said about people in the internet being easily offended “snowflakes” these days, I’ve noticed that people who claim to being cancelled actually have (1) said something divisive or not clear enough so as to cause a public uproar, and (2) are simply taking on a victimhood mindset when justifiably-outraged people react to their content, instead of coming to terms with the fact that people are allowed to disagree with them when they post controversial thoughts on social media.
What you can hear, are the near misses, like people employed by an institution that took a vote about whether to fire them, and decided that it wouldn't. Or people who took a hit and recovered, like they were fired but found another job later... maybe a less paying one, but they can still afford an internet connection.
But this is just the tip of the iceberg. I assume that for each person who spoke bluntly and lost or nearly-lost their job, there are at least ten people who decided to keep quiet.
> said something divisive or not clear enough so as to cause a public uproar
That's in my opinion quite sad that being "not clear enough" is a punishable offense these days. In the past, people were less hysterical... well, probably the hysterical ones were just less interconnected.
Like, seriously, if someone says a genuinely stupid thing once, would you fire them from their job? This seems completely crazy to me. Did we completely lose the ability to forgive / forget / ignore?
> This means the conversation falls back to tribal bickering and shouting
You are right that this will always happen, but I do not think it will be the dominant interaction. One of the reasons this interaction comes up so much is that platforms optimize for it - bickering and shouting is engaging. It gets clicks and eyeballs and shares. Jerry Springer was a popular tv show (and good for that team) but it's wrong to say that we can only make TV like Jerry Springer.
There's a great deal of research in the UX / Social Science world on "affordances" - UI features basically. They're the nouns and verbs that we have in digital spaces. Often the research is very basic but it is out there. I think social media companies could do a lot better with the affordances they give to people. They could be more playful. They could be more free form. They could slow down conversations where people are rapidly tagging each other and sentiment analysis suggests they are fighting. Etc, etc.
The human brain is a very complex thing and one of its primary purposes, as far as we can tell, is to be social - the thing you say isn't a task. I don't believe being social isn't a task. I just think we're very bad at helping our brains be social (a thing they evolved to do for millions of years) in a medium we just invented maybe fifty years ago. We can get better, but only if we try.
It's as simple as: use of recommendation algorithms + forums at scale.
When you do this, people are automatically shown things that piss them off (outrage has higher engagement).
The bigger the scale, the worse it is.
Once you get past a certain scale you could get the same effect without the algorithm, but the algorithm makes it much worse.
Are you sure about that, or do you just think so because more social science is done on Twitter due to their relatively open API, and conveniently short post format? If you're sure, I think you should shore up your claim with evidence.
The thought shaping at scale capability of Reddit is massive, and seems to go mostly unnoticed as everyone is laser focused on QAnon and conspiracy theorists (which quite conveniently are very common topics of discussion in these threads).
Let's take look how Hacker News is designed: A discussion platform that utilize tree structure to keep discussions structured, easy to track and on point.
There are many different kind of designs, the flat-list one for example, which causes less structured and often ineffective discussions.
The problem of social networks is not just due to social factors, software design plays a key role.
You know, most physical real-world bars/clubs, which are also social places, couldn't generate that level of toxicity. Even the toxicity on Reddit is no match for the top social networks. Those systems are what I called "moderated-islands", on each island, there are people who wants to keep things in check.
On the social networks however, the people who checks things are the same people who post the content, and moderation is done directly under the content they've posted, there is no "the third eye". Then, you introduce the feature that prevents your users to from structured discussions, and then give them a "How popular U R" counter ...
It's more complicated than "It's just social".
Provide the feed based on post time, allow you to agree or disagree to view hidden posts due to friends votes (down and up).
That's it. I useto enjoy FB(like 10 years ago), until I lost the ability to scroll through a time based list of posts.
That seems more like it. What does that find? Ah, here we go, "COVID-19 Vaccine Side Effects" - a group with 66k members. And lots of juicy, fascinating stories of bad things happening after someone got vaccinated.
And 10 minutes later, even I, who knows that in a country that has delivered almost 100M shots, shots that don't protect against stroke or heart attack (or eczema or whatever), there are going to be a lot of folks who get the vaccine and then have something bad happen to them can't help but feel at least a little more worried about the prospect of getting one of those shots. (I don't have to imagine the effect of these stories on folks with a weaker stats background - I just have to talk to my relatives...)
And that's something that's not necessary, but is extremely lucrative for FB. And that's a problem for anyone who likes living in a functioning democracy.
Even in your example, you had to go digging pretty far for that group -- you didn't accidentally happen upon it. If you people are motivated to find bad information, they will. That's the nature of the internet. If you don't want people to be able to gather and share information, then unplug your Ethernet cables because that's what the internet is all about; FB or no FB.
In real life you won't go social with others debating on a heated topic, maybe we shouldn't do it on social network, slow down a little bit, we don't have to change the world in one day.
1. They cannot be monetized via anything but ads. Sure a few people will pay for access to a social network. But the majority will not pay $0.01 because the friction of the transaction will prevent them. Moreover, an SN operator can make a lot more money from ads than from subscriptions, at least until proven otherwise. As a result any useful value that an SN provides will quickly be overtaken by more and more intrusive ads.
2. General purpose social networks are relatively boring. I mean how often do you actually hang out with a proportional representation of the general public? No, we want topic or interest based social networks. We want to form consumer tribes, political tribes, sports tribes, etc. We want to talk to people from our school, work, neighborhood, yoga practice, etc.
3. All that is overshadowed by the fact that most people straight up don’t have anything interesting to say. Think of the average person. Do you expect them to wow you with what they are working on even once a month? Chances are, most people will post pictures of their lunch or repost a stupid meme that they finally discovered. Twitter sort of solves this by letting you follow leaders in their fields, but still the SNR is so low that it’s really difficult to find meaningful interactions and unless you are one of those leaders nobody gives a crap about what you shout into the void.
The only thing I can see being a path forward is small communities forming their own very limited in size networks with some common authentication platform to help people jump between them. Discord is sort of that, but far cry from a sustainable model.
Starship Troopers (both the book and the movie) hit the nail on the head... "Something given has no value."
The thing is that if your social network gave you more value than $10/month, you’d sign up for it, no? Remember when texting on your mobile phone was a separate plan that cost about $10-20/month? People fell over themselves to sign up for that because it gave them access to a large social network, one that consisted of people you actually meet in real life and that are presumably worth keeping up with. I don’t use Facebook because nobody on there really is worth my time to keep up with, and I can’t bring myself to try to curate my feed/friends list because that’s just work, all the while FB will keep polluting it with ads anyways. Instagram is mildly more worth it to me because you can actually find some meaningful connections on there as a photographer, but I suspect that’s 100% temporary.
TANSTAAFL is from The Moon is a Harsh Mistress, basically a libertarian manifesto. Heinlein had range. Add Stranger in a Strange Land, the hippie manifesto, and you have the trifecta.
That asymmetry enables toxicity. And volume will make it a matter of time until arbitrary toxic things eventually show up.
Participation must require more committment than opening a free account.
I wonder how those free accounts without any accountability get any credibility.
I don't know about this. It may well be that only being ad driven can monetize them to $billion+ valuations. But I don't know if it is for sure that you can't have a more modest, but still very respectable business worth in the hundreds of millions of dollars by monetizing in other ways.
It should also be clear that social media's monetization strategy isn't just about ads, it's about data farming to serve highly targeted ads. They're not just selling out full page ad spreads and selling classifieds like in a newspaper. Those would be perfectly fine monetization plans that just focus on self-reported areas of interest and some very basic geolocation. There is some incentive to drive engagement to get eyeballs but it's not nearly as strong because they just need you to log in and hang out. They don't need you to interact constantly to collect data on your emotional state and receptivity.
This could be done by the ISP hosting an instance of the service that can inter-operate with other hosted instances on other ISPs. They already do this with email and many of them did with usenet.
Back when ISPs provided access to usenet, they never filtered what was posted and what you saw was under your control.
I tend to think of IRC as sort of a proto-social network (channels are kind of like tribes - there's usually a topic that is of common interest for those participating). It has existed for a long time without ads or even monetization.
IRC has no power to shape national discourse, international politics or even quotidian humdrum discussion topics because it has a high barrier to entry, low user friendliness and no additional features.
Personally I wouldn't have it any other way: IRC is perfect as is, but it is not a social network that has ever gained enough steam to devolve. It's like saying your family blog has a great comment section, therefore all comments sections can be great without devolving: that's not even universally true for family blog comment sections.
But, except for that, I totally agree with your comment. It did never have the power that social networks have today, and as soon as more interesting tools were available, most people moved on.
Page loads, reloads, dynamic content, rich media and more does add up to amazing amounts of data. As the user base grows, monetization or financial donations becomes a requirement if such a platform will be sustainable. A lot of processing and data transfers "need" to happen to simply read a short text message from a friend.
How could we design social platforms as streamlined as IRC? What technologies do we have? HTML works. How do we keep the online experience and rich media we are accustomed to today while going for an affordable architecture.
Solving the ads and monetization issue should be tackled from multiple angles. Not just asking where the money will come from. I believe more could be discussed about creating sustainable, lightweight solutions as well.
I've been trying to open the network up to the public more, but I'm starting to realize that making the communities more insular and invite-only had been more successful - but that's very hard to scale.
The site is https://podaero.com - always appreciate new users.
In other words, the meetup model might work for these kinds of communities.
What are you doing to market your project, and have you had an easy time getting users?
As you can imagine, it's hard to crack the chicken/egg social platform problem, so user acquisition is slow going and I'm completely bootstrapped. I'm in it for the long haul though (it's been over 8 years now in the making).
Kind of a tangential thought - it would be kind of cool to have crowdsourced 'prediction markets' for the tv shows - i.e. people guess what's going to happen in an episode or season.
Aside from that my motto has been that if you say something that doesn’t exist should exist, it’s now on you to make it happen or to find and fund someone who will. Organize an event using this service or another, advertise it to a few friends, make sure everyone has a fun time, they’ll spread the word for the next event for you.
What if ads were banned?
If you just want to talk to your buddies or your church group, yeah you don’t need an ad supported network for that.
See how many people pay crazy money for Bloomberg terminal just to be part of the in crowd.
A social network could work for a few bucks a month I think.
Imagine the implications of actually believing this, because if you thought hard enough about it, you probably don't believe it. For example, if you truly believed this, you must think most people straight up don't have any friends. I mean, why would anyone want to be friends with someone who straight up doesn't have anything interesting to say?
Like, do you think the average person, hanging out with their average friends on a regular basis is just mind-numbingly, dumbfoundedly bored? Why do you think they keep hanging out with all those average people? You think they derive zero meaning from their interactions with anyone who isn't a "leader in their field". The dim view you have on humanity blows my mind. There's meaning to be had all around for these average people.
This is exactly why average people hate the proverbial elites. You (ostensibly elite) view them as worthless plebs whose social interactions with their friends are so worthless, they don't even deserve the chance to share them online. Only so-called leaders in their field should be allowed to interact publicly online. Their interactions are the only ones with meaning in your view.
When I'm on FB or Instagram or Snapchat, and I see the plenty of average people I know interacting, I see all sorts of wonderful, meaningful human interactions. Today, my old high school math teacher posted a joke image of the side effects of the COVID vaccine which included hugs from his grandchildren. His son replied about how he can't wait for him and his grandson to be with him again and many of his friends commented about how touched they were. Who gives a fuck if it's a cringeworthy, stupid meme they discovered? It was a wonderful expression of humanity from completely average, nobodies.
There's so much wrong with social networks, but the plebs having no meaning and nothing interesting to say isn't one of them.
I firmly believe that technology like social networks can bring people closer together and enrich lives of everyone, from luminaries to ordinary folks. And I believe we should build better networks. But if I throw a dart at a completely random Facebook post and read it, there is a very good chance it’ll be a lot less interesting than a random post from e.g. Neil deGrasse Tyson or Neil Patrick Harris or Obama. My point is that having these huge global networks doesn’t actually achieve the goal of furthering friendships or increasing the quality of content the participants read.
Uh, what? No they aren't. If I shared something on FB or Instagram, it would be shared exactly to the people who I'm friends with on FB or follow me on Instagram. I, and I'm willing to bet really large amounts of money at favorable odds to you most other people, don't have anything close to 10K people in either of those categories. Weirdly, despite your favorable comparison, and in fairness to your point, Twitter is much closer to being that megaphone.
> This is why your feed, while having some gems, is also likely full of reposted memes, personality quizzes, and other retweeted/reposted content.
Actually it's almost entirely not this. It's mostly boring baby photos or engagement photos or posts about middling shit in most people's lives. I wouldn't be surprised if FB downranked that type of content a lot. On IG, it's literally none of that. Mostly baby photos and photos of little trips and outings people do.
> If you have no actual special connection to the person who is posting the meme/quiz/repost, it is just noise to you, no?
Why would I be connected to that person???
> there is a very good chance it’ll be a lot less interesting than a random post from e.g. Neil deGrasse Tyson or Neil Patrick Harris or Obama.
NDT: https://twitter.com/neiltyson/status/1371154155931377666 Just as interesting a meme
NDT: https://twitter.com/neiltyson/status/1371147864361996293 a meme joke about pi. So profound.
NDT: https://twitter.com/neiltyson/status/1371139056218742784 Did you know pizza is known as pizza pie and pie and pi are homonyms? Wow, thanks, Neil! Glad a leader in the field like you could hook me up with this interesting content that no average person could!
NDT: https://twitter.com/neiltyson/status/1371121692064374785 Normally, I would need a calculator for this content, but NDT has got my back on Twitter.
NDT: https://twitter.com/neiltyson/status/1370797127459278856 cute cows are cute, also pithy political statement. An average person might be able to post a cute animal OR make a pithy political statement, but BOTH!?
NDT: https://twitter.com/neiltyson/status/1369144426845663234 buy my merch! No average person could ever have merch! He produces great content like this on a weekly basis. Even if an average person was able to have merch, could they post about it weekly??
NDT: https://twitter.com/neiltyson/status/1366730477642649603 Faux deep statement with a picture of space. My grandma literally posts shit like this all the time, I'll let her know she's a leader in her field.
NDT: https://twitter.com/neiltyson/status/1366097324276027393 LOL dogs smell butts. Classic content no average person could do.
NDT:https://twitter.com/neiltyson/status/1365764623820029959 LOL DAE dogs shed on furniture!?
NDT: https://twitter.com/neiltyson/status/1363840108861005829 Same vacuous quips my teacher friends post when they're feeling down.
NDT: https://twitter.com/neiltyson/status/1363159290601758723 Seriously, it's all memey bull shit. I'm not even being selective here!!!
Let's do NPH.
NPH: https://twitter.com/ActuallyNPH/status/1364362084985823234 Thanks for saying nice things about me! Average people are never thankful for things in their lives!
NPH: https://twitter.com/ActuallyNPH/status/1355157217478123524 Just like MLM shit people post.
NPH: https://twitter.com/ActuallyNPH/status/1352751015121063937 Shameless self-promotion. Average people don't have personal brands that they self-promote. Just leaders in their field like NPH. This is the vast majority of what he posts. Thank GOD I can get this on Twitter by following leaders in their field instead of FB or Snapchat where people share stupid shit like their vacations with friends.
NPH: https://twitter.com/ActuallyNPH/status/1328787438232276993 Last one because I can't stand how ridiculous your lionization of these people is over the average person. Here he recommends a book, which of course average people, on account of being illiterate, could never recommend!
I'm not too optimistic about this and I don't think it will see widespread adoption, in particular not without any kind of legal framework behind it.
When Berners-Lee talks he still has essentially the early web mindset. Build it, and individuals will come voluntarily because it looks nice and we'll make friendly communities... somehow. I think if that was enough Mastodon would have already replaced Twitter.
For some of us, it has. ;)
For others who are fine with Twitter, well, there's still Twitter.
I realized earlier today we're still in the era of AOL. Only now AOL is called FAANG and instead of snail-mailing you plastic discs, they pre-install on all your hardware. The people who either prefer the corporate ecosystem or simply don't know any better hang out there while the wild west waits just outside.
That decelerates the trolls and doesn't harm the others. Entry must be possible for laypeople, however, and can be, if a few tools (like my https://demo.mro.name/shaarligo) exist and improve.
And self-host. Very few people burn their own garden.
All it takes is an adult approach where candy isn't free.
The problem with social networks is moderation. Moderation can only really be done well by users; not by some corporate top-down approach like Facebook, Twitter, et al have been trying.
Here's what I think the future of social networking will be:
https://www.quora.com/What-is-the-future-of-social-networkin...
I suppose it's fair to say that there isn't as much outright malicious disinformation on my LinkedIn feed, but to me it's a distinction without a difference -- crap is crap.
edit: And at least when I see crap on Facebook, it's usually my own fault to an extent: it's almost always from someone I follow directly. LinkedIn seems intent on populating my feed with almost entirely stuff from people I don't know, about subjects I couldn't care less about, just because one of my connections "liked" it.
LinkedIn exists to benefit Microsoft, not you.
Limiting how people can interact and reward people for reacting positively, is a great 1/2 punch way to prevent the nastiest people from entering into a positive feedback loop.
Yeah, it's not as free, but I think we've gathered enough evidence that freedom (in this social media context) causes more problems than it solves.
I think what we really need is exclusivity and accountability. So you can't just be messaged by random people, and if people messaging you are being assholes they will at minimum be kicked off your network. I think this is similar to how "real" friend groups work, how they worked before social media. The main problem is it's hard to find new people.
My problem with accountability is that it doesn't seem to matter. Facebook users have shown us that even linking terrible content back to real people won't stop the real people from sharing those terrible content. Anonymity (or its inverse) isn't the answer.
Has anyone tried putting up a paywall to use comments? I feel like surely someone has, but I can't think of an org off the top of my head.
I don't think the goal should be to maximally facilitate social interaction anymore. We've proven there are enough bad actors that it poisons the well.
So, given the above, how can we still enable some interaction? How can we get the value out of crowdsourcing without creating an opportunity for toxicity?
The issue in my view hinges upon making it easier for people to choose who they interact with online, including tools for delegating those decisions to zero or more third parties.
We'll need a good decentralized identity system first.
The other problems with social media are either not important to most people, or solved by other services already.
E.g. you post a picture, and I comment on it. Doesn't mean your mother should see the comment too.
Among close friends, it would be obvious most of the time. But most of our social media contacts are not close friends, and it's not clear what these reactions would mean.
Facebook didn't cause or perpetuate the situation in the USA, it simply reflects it, as any communications tool.
Let me tell you: Facebook sadly has no monopoly on partisan lies and associated screaming, or rampant misinformation.
I'm disappointed that TBL, who seems to be very smart, falls for it so easily. Like when he mentions "feminist bloggers" - these people get toxic replies because they write toxic articles and messages (basically "all men are assholes", so naturally some men get angry about the insult). Haven't experiments in fact found that men receive much more hate messages than women, they just don't make such a fuzz about it?
I don't think TBL actually advocates for censorship (except with his "curation" idea) when he asks for better controls. Good muting, blocking and reporting features go a long way towards making social networks usable.
Facebook offers shitty controls, and reads and influences our conversations. Unlike cellphone carriers.
What is required is to shift focus from social networks to the society. Removing hate, bigotry, evil and social factors that induce crime ought to be the focus.
A society cannot be reformed if it is corrupt at its core, no matter how many controls you put on public communication systems.
The problem has to be addressed from within, not from outside controls.
Are we supposed to celebrate whoever created the plumbing or the interface? After all, Tim is not responsible for the user adoption and for all the user content created out there. If it were not for WWW it would have been someone/something else and nobody would know Tim right now.
Marshall McLuhan, using his theory of media, was effective at diagnosing the downsides of social media decades long before it became the dominant medium through which our worldviews are shaped.
I don't think Tim, nor anybody else in the tech industry* was listening when McLuhan claimed "peer to peer electronic media" would bring out our tribal impulses, but here we are. The internet is a bunch of identity groups, each working with distinct sets of facts, all of them suspicious and distrustful of the others, who enjoy beating on one another for sport.
Of course, humans are inherently tribal, and civilization is a hack. For it work we have to suppress, and find safe outlets for our basic tribal impulses.
*Zuckerberg has claimed to be a McLuhan fan, but if he actually understands what McLuhan was saying, then he's straight up evil.
The greater the mental charlatan, the more definite his insistence on the wickedness and weaknesses of human nature. Yet, how can any one speak of it today, with every soul in a prison, with every heart fettered, wounded, and maimed?
John Burroughs has stated that experimental study of animals in captivity is absolutely useless. Their character, their habits, their appetites undergo a complete transformation when torn from their soil in field and forest. With human nature caged in a narrow space, whipped daily into submission, how can we speak of its potentialities?
Freedom, expansion, opportunity, and, above all, peace and repose, alone can teach us the real dominant factors of human nature and all its wonderful possibilities."
- Emma Goldman, "Anarchism: What it Really Stands for" (1917);
I can't imagine how different (and more socially vibrant) life was 40 years ago.
I feel like philosophy has been debating this issue for millennia.
I think that trying to look at it from the perspective of innate goodness or badness of an individual really misses the point.
In my world view, everyone basically has the same motivations[0]. You can understand this in terms of what philosophers call "negative freedoms"[1] - freedom from hunger, freedom from pain, etc (the linked wikipedia article doesn't fully convey this sense, but it's what I learned in philosophy class). People, basically, want to "eat, drink and be merry".
Individuals naturally adapt their behaviour to satisfy these freedoms - for example, you get a job so you can be free from hunger and free from danger, among other things. Of course there are positive freedoms and aspirations too, but the basic stuff, like not being hungry, or wanting your kids to do well, is pretty much universal, and deeply informs my view of the world.
Anyway -- as an individual, pursuing your desire to eat, drink and be merry seems like a pretty harmless way to go about life. But if that individual joins a group, and some people in that group tell them that the reason they can't get a job is because <insert group here> has <insert cause here> - then their behaviour will reflect that group dynamic for all the reasons we already know, and that behaviour will become self-reinforcing within the group itself. You can see how this has played out in US politics recently; it's very obvious and I don't think I need to give examples.
The problem of groups comes about, in my world view, because groups are completely different from individuals. It's far harder to change the opinion of a group, than it is to change the opinion of an individual. It's really a scaling issue; once you get a group of similar people who reinforce their own beliefs - for example, that certain other racial or ethnic "out-groups" threaten their ability to meet their needs - then they are going to feel justified in taking action against those groups.
There is no question that the actions of those individuals and of that group are bad, because in my worldview everyone is equal and nobody deserves to be hurt when they are just trying to "eat drink and be merry". But the perpetrator, too, was just doing what they thought was right, in order to achieve the same goals. That they were misguided is entirely different from them being "bad people".
That doesn't make the actions of the individual OK, but it obviates the notion of people being innately bad. To be clear: if they do bad things then they still need to suffer the full consequences of the law for their actions (the law's role is not to punish people for being bad, but for doing bad things). It's never OK to hurt people for any reason. But it's senseless to say if an individual who likes puppies, looks after their kids, likes a beer after work, donates 10% of their income to charity, and is also a racist, is "bad", if the latter attitude is the accepted behaviour of the group that they belong to.
I feel like I need to be super clear here: my world view is that racism is stupid; racists do bad things; and the leaders of racists groups are scumbags. But humans learn by observing, so a group member who's a racist is, at worst, generally redeemable.
So, in my world view, it makes no sense to rank people in general in terms of their individual goodness or badness, because I suspect that most people are basically good; I know this because I can walk the streets of my suburb and not get stabbed. It certainly *does* make sense to rank group leaders in terms of good or bad, because groups ossify group behaviour; the very behaviour of in-groups that I pointed out above can be weaponised by unscrupulous people, which includes the obvious historical figures and many, or perhaps most, politicians.
So - to sum up this overly long post - I believe that individuals are almost universally good. Groups, on the other hand...
[0] Clearly, this is not absolutely true. Some people have pathologies which maybe make them unlike this - wanting to harm people etc - and some people are willing to hurt others for personal benefit. But in my world view these people are very much in the minority.