The headline claims it's "no accident", but couldn't these properties evolve out of a lot of A/B testing without the product managers or engineers realizing the broader behavioral implications?
Sure. But that also doesn't mean it's an accident. They're trying to increase engagement, and what do they have at hand other than addictive techniques?
> ...and what do they have at hand other than addictive techniques?
Reaching for addictive behavioral models is like reaching for management policies when the hard work and long-term investment into leadership is more sustainable.
To directly respond to your question, they can for example move to a facilitation model; engagement is encouraged through longer discourse, or through editorial staff whose job is to summarize viewpoints/trends/patterns that quickly bring newcomers up to speed, but they do not moderate.
Another possibility is an aspiration model that shows the small steps and decisions it takes to succeed, rather than orient towards the highlights reel view favored today. Say someone you admire and follow wakes up at 3 am to go work out: the social media connection time-shifts to your timezone, corrects for your distance to your gym, and encourages you via your smartwatch to do the same, in realtime. They eat exactly 100g of oatmeal for breakfast; the social media connection corrects for your body measurements (lean body mass, gender, height, etc.), and says you can eat 90g to follow your aspiration person.
Yet another possibility is community-wide goal setting and pursuit. Engagement through groups creating their own layouts, platform-based applications, etc. to both decide and pursue their goals.
There are more models to consider. I posit they are not pursued by FB because they decided they cannot monetize those models under their current ad-driven monetization model, which in turn is governed by the capitalist (though not necessarily free market) model they operate under. Relax some or all of those constraints, and the other kinds of engagement models might emerge?
It's not really an emergent property but anything driven by selling attention has been engineered and tested to draw attention reliably. From print media to radio and TV the formats are chosen in a way that boosts engagement. From publishing short stories in serial in the 19th century to closing a local news segment with a teaser for a breaking story, these products are designed to be habit forming.
This is a good topic -- I mean, in the sense that I believe the appeal of social media is firmly rooted in our desire for social connection and validation. But that doesn't mean that the design of buttons and interaction-loops can't meaningfully add to the addictive nature. That said, I thought the OP's examples were a little flimsy without citing evidence (that the "few moments to load updates" is deliberately designed around psychology of intermittent variable rewards. I mean, sure, but the time to refresh for updates is not non-existent, even with high speed internet).
Examining the simplification of choices to binary is a nice example, IMO, of how to make social media feel less "work":
> Following the introduction of Facebook's like button in 2009, YouTube moved to a binary like/dislike format in 2010. Instagram launched that same year and came ready-made with a Like function shaped as a heart. Twitter adopted this same heart-shaped system in 2015, while, in the years since, Silicon Valley has come up with a multitude of new ways to gamify our need for social validation.
I've never really liked FB's multiple emotions, though that is partly because of the interface design in which it just takes longer to register one of the other non-Like emotions. Netflix's move from stars to thumbs-up/thumbs-down felt controversial, but I'm pretty sure I've "thumbed" more things in the past month than I had made star ratings in the past year, simply because it was easier. And the data that Netflix provides me in terms of potential movie appeal -- a likelihood described as a percentage, versus a five-star scale -- feels more straightforward, and doesn't seem to significantly differ in accuracy from what I remember of what Netflix used to recommend for me. Roger Ebert's print reviews used a 4-star scale, but his Thumbs-Up/Down with Gene Siskel was still very popular and widely influential.
On the topic of social media engagement design, I would love to hear more about the recent Facebook feature that prepends my name before every notification. e.g. "Dan, you have 32 new notifications and 3 pokes today." versus "You have 32 new notifications and 3 pokes today." It's such an easy an obvious feature to implement, I wonder how long it was in the making? e.g. A/B testing, psychology studies, etc. Can't say it's really worked on me except to make notifications feel more of a nag but I'm likely an outlier in that I generally rarely visit FB.
On the topic of "few moments to load updates" while I also could not find anyone claiming to have solid evidence on this practice, FB having a history of being intentionally being anti user with it's android app is pretty well accepted. I wouldn't put it past them, though I suspect it is obfuscated in the form of javascript bloat. Anticipation plays a big role in addiction.
This concept is much bigger than a pop science book.
For example, the article quotes the author of a different pop science book on the same subject; quotes a different source on ethics; and references intermittent variable rewards which were pioneered by BF Skinner almost 90 years ago.
Yeah, I was just wondering out loud because it's a source "of the times" from people building these sorts of addictive products knowingly. I just read the book and was surprised it wasn't used as proof that people are knowingly doing it.
Scanning the article one more time, it seems pretty limited in focus to "like" buttons and social validation. You could probably write a whole book on the other types of products and devices :)
This article also uses manipulative tricks to maximize your time on page. Notice that each paragraph is relatively short and approximately the same length. This is done purposefully to reduce the cognitive overhead of reading and makes it easier to keep going from one to the next, always with the hope that maybe the next paragraph will deliver some key insight.
In disagreement with the parent's snarkiness, I would argue there's a fine line between efficiently structuring written information to convey important news in a time sensitive manner and using bright red numbers to keep users swiping your app all day. Especially when it comes to messing with peoples self-worth and validation!
"You may have won $100! Watch these 5 advertisements to find out!" is gross but palatable to me.
"People might like you! Refresh the page and view another ad to find out!" is appalling.
In the sense that it exploits only the reader's poverty or greed, and even the most credulous will soon tire of watching ads, close the page, and be wiser for the experience.
Whereas "People might like you! Refresh the page and view another ad to find out!" cruelly exploits a desperate emotional vulnerability. And there is no end of it in sight.
Have you seen the Imzy submission [1]? People have argued that they failed because there was no content on their front page. Why have who chosen to require logins?
>1. No front page (for lurkers): They force users to join to be able to view content. The 90/9/1 rule is real, ESPECIALLY in the "community" space. 90% of users just want to lurk and that's just the nature of the internet. It's probably the same here on HN. 9% will participate in some ways, voting, etc. 1% will submit and comment, etc.
This is a very valid point. We are considering adding an option to view conversations without login, but we feel that this should not be the default behaviour (there is an expectation of privacy) but would only be used were it the default.
I've thought about this issue, but it seems like any type of advertising or program that refuses to respect boundaries and pushes psychological buttons is bound to win over a product that does not.
This is true if the market is an inert resource, ripe for exploitation. However, the market is beginning to realise that social media is at the moment a bad actor, and that their attention is being traded as a commodity. This realisation will bring a massive exodus towards services which respect users and language.
I'm already looking at setting myself up with a private Mastodon instance for my friends, I never liked the idea of funnelling my life into a service I don't own and I don't think my friends do either.
Mastodon is great, but its nature makes it inaccessible to those who would benefit most from open discourse and conversation: the layman. It's got a high barrier because it's aimed at the more technologically literate.
It's no harder to register an account on any public Mastodon instance than it is to register for Facebook. In fact, it is arguably easier, since Facebook often requires phone number verification nowadays.
That's true, but the distributed, decentralised nature is something that the layman is often not interested in understanding, and so can turn people off the network. Neither does it have a cuddly name!
If I understand correctly, this is thoroughly different from "social media." In fact, it looks like "connections" are all but irrelevant - only "conversations" matter. Of course, conversations are richest when you have an understanding of the other participants background and depth of knowledge, so you can explore, test, debate and grow your own knowledge, or adjust your beliefs.
My impression is that Lyra targets a niche of individuals that "enjoy conversing." I'm not sure if there's a nexus, a central target topic of conversation (like a sub-reddit would have) or a connective tissue such as real-life familiarity (like a friends and family social network would have) to hold your users together.
Can you elaborate on who your target audience is, and how they will be attracted to the site, or what will maintain their interest (while avoiding the trap of making their engagement addictive?)
Connections are relevant in the form of groups, which are private to you: you use them to set conversation audiences and control whose conversations appear in your news feed.
Our target audience is those who want to converse online without the restraints and commitments of social media services which aim to addict rather than to support and convey language.
We don't aim to enable the discovery of interesting links - Reddit has this completely down. But it's no good for conversing in a focussed way with people you know.
Letters and email link people internationally in ways that are impossible on Facebook or Twitter. When you receive an email, you read it in a focussed space without distractions; and you're aware of the fact that it was personally sent to you as opposed to broadcast to a large channel. Narrowcast will always be a personal experience than coming across a message delivered by a mysterious algorithm.
Lyra does have a public news feed in which conversations are viewable to all users. If you add people to your groups and set the feed to view only these, it's the exact equivalent of Facebook.
What we don't do is offer particular spaces (hashtags or subfora) because these instantly offer a target for harrassment and abuse, and require moderation. Lyra's design makes it trivial for people and topics which are vulnerable to abuse, to control their reading and avoid it.
As for engagement, we are reliant on guerilla marketing and links with aligned groups. And as for maintenance of interest - Lyra is a very pleasant space for written conversations. By default, (configurable) you get an email when someone starts a conversation with you or replies to one of your messages, so you're informed of activity. Conversations between 2 people or small groups are a lot of fun.
While the concept seems really interesting, what do you mean by "open" in this context? For example, are you able to export your data? (I wasn't able to find anything about that on the site.)
By "open" we mean that the platform does not discriminate against any particular groups (for example, by requiring real names) and that its feed algorithms and behaviour have a clear and simple meaning.
I'm building tool[1] for myself to organize book reviews, my essays, quotes based on same principles and inspired by writings of Joe Edelman and Tristan Harris.
Lyra looks really nice
[1] https://montaigne.io/
For those who want to get rid of facebook addiction without losing facebook account: just unfollow everybody in your "news feed". It is not visible to anybody, you're still friends, but stuff just stops appearing on your fb page. In a couple of weeks you'll end up with empty page, and facebook will disappear from your life, because there's nothing to see there anymore.
I built a chrome extension to help me with this addiction. It replaces the distracting parts of social networking sites (like Facebook's News Feed) with a todo app to keep me on task whenever I look to these sites to procrastinate.
The app unlocks the site for a configurable amount of time after I've completed my tasks.
Also supports Hacker News, Reddit, YouTube, Twitter, and Product Hunt.
It has helped me a lot with a. Recovering lost productivity and b. Feeling less addicted to these sites.
Now if someone can figure out a way for Facebook to rewrite their site in pure WebAssembly, they won't have to worry about pesky programmers like you breaking people's habits.
That's the most vexing thing I've read in a while. I manage Facebook addiction by not having the app installed on my phone and only using the browser(which works just fine). If they ever compiled their site to WebAssembly, that would be the end of them for me. The whole idea of WebAssembly is simultaneously awesome and scary.
I guess I'm new to the whole WebAssembly thing. Could you please elaborate more on that statement? If FB were to do that, how does WebAssembly stop you from doing what OP suggested?
You can't modify or block the newsfeed if they don't even have a DOM. Just feed raw pixels to the screen via webasm. And they have some incentive to do that because like many sites, they are in a constant war against scrapers and Adblock.
Theoretically, it doesn't. Practically, it would be a complete nightmare, as you'd have to decompile the source code into a high level language, then come up with some kind of (proper) hack, and run a code insertion tool in your browser to reinsert the changes. And they'd probably break any time Facebook altered their code at all, and moreover, FB could use a polymorphic compiler to make such things impossible. And finally, decompiling the program would probably be illegal if it was to circumvent the intended use of the site:
I believe the assumption is that WebAssembly code wouldn't run on the typically browser API accessible DOM. Therefore chrome extensions, which operate on accessing markup via JS would now be rendered obsolete.
That is awesome! I have been using News Feed Erradicator for a while but that is next level! Too bad I am using Firefox ;)
Also, recently I realized that when I in nervous procrastination mode I also check GitHub too much to look at the stream of updates and "someone starred ..." stuff, I have looked for disabling the stream but could not find it... I guess I could try play with the Stylish plugin to make it disappear or sth.
Admittedly, sometimes I wish I could remove all "points" and "numbers" from the web, even on HackerNews. There is this quick brain sugar rush when you see these numbers rise, but just a minute later I feel empty, sad, and dirty for having this narcissistic automatic response.
This is one of the reasons (the other ones go around privacy and tracking) that I don't do Google Analytics nor any kind of access.log aggregation/analysis on my websites. Maybe this is a stupid rationalization, but I'd like to think that we value stats too much and that they distract us from doing better qualitative design and analysis...
I'm toying with the idea of having the Todobook extension replace all Like/Comment/Upvote counts, etc with the number of tasks you have in your todo queue. I think the Likes count is toxic and encourages really weird behavior.
I don't like how Reddit only shows the user top posts for past 24 hours or past week. It means one can't easily not look at it today and see what one missed today tomorrow.
I'm skeptical about some of the examples provided. (Do we really know that Instagram's 3-second delay is gratuitous?) But I don't doubt that social media is pushing people's mental reward buttons.
This seems to be happening not just on social media, but throughout the consumer technology space. The most obvious example is how video games skinner-box the customer, but I've also become cynical of how my industry (television software) is driven by consumer addiction. When I started working on television applications years ago, I saw it as an opportunity to help people experience art. Now I see people spending every spare moment glued to video, and I just feel like a drug dealer. :/
Getting likes is addicting, sure, but scrolling through my newsfeed and seeing autoplay videos with obnoxious text banners in impact font ('me af rn crying-laughing-face-emoji fire 100 100') has made it so that I'm very annoyed with social media.
edit: Personally left Facebook a little while ago. It had reduced in my time to the odd few minutes scrolling threw the news feed. It just did nothing for me. I don't miss it.
> Getting likes is addicting, sure, but scrolling through my newsfeed and seeing autoplay videos with obnoxious text banners in impact font ('me af rn crying-laughing-face-emoji fire 100 100') has made it so that I'm very annoyed with social media.
I have never understood this complaint. Why don't you just unfollow those people? How many of them could you possibly know? This seems like a problem that can't exist for more than a week or two if it actually bothered you, since unfollowing someone when you see low quality posts like that will clean up your feed pretty rapidly.
(Note that unfollow is different from unfriend on Facebook)
I don't have a facebook account. I had a twitter account that I used for a news feed for many years. Meaning I really didn't tweet, I just read other people tweets. Long story, but my account got suspended. Then it occurred to me that maybe I don't need it any more. It's been 20 days, I really haven't looked backed.
This is a side effect of their dependence on advertising. It's spillover from the traditional media business model.
The consumer doesn't receive any real, tangible benefit from the product. If they did, they would be willing to pay money for it. The only way social media, and much traditional media, can profit is by collecting attention and selling it to advertisers, so that's what they are optimized to do.
It's an interesting thought that, if the users really got a value out of Facebook, they would be willing to pay for it. I think people do, but if Facebook asked everyone to pay even just a penny, they would instantly lose hundreds of millions of users. There just isn't a way to charge users directly for occasionally useful products.
Almost all of the services people actually use Facebook for were things that they paid for in the past. It's certainly harder now that these things are "free", but I think there would be a market for a Facebook clone that said "Hey, just give us $5 a month and we won't sell your data or bombard you with ads".
Let's see, keeping up with your friend? That used to be the telephone, for which people paid by the minute - an hour long call with your friend or family in another state might run $5, every time you wanted to talk to them. Or you could mail a letter. Slower, but still costs money for paper, envelopes and a stamp.
Wanted to share photos? First you had to buy a camera. Then you had to buy film. Then you had to pay to get the photos developed (extra for duplicates). These are clearly things that are (or at least were) worth money to people.
But a stream of random snippets from people you haven't spoken to in 20 years, asinine clickbait, and political rhetoric? Yeah, nobody wants to pay for that.
I find these sort of articles annoying. I've not read the book referenced in the article, but I've seen some of Tristan Harris and I think he's a smart guy but stuff like the 60 Minutes interview plays into him looking like a hack who is far more interested in self-promotion than in the actual topic.
It just doesn't compute that "knowing what drives users" == "addiction manipulation". Yeah, there are slot machine-style techniques that you can use in those areas, but they don't work well or for long and users grow to resent them.
But the stuff that keeps getting brought up, like "people on Facebook are desperate for validation and connection" makes it sound like users are forced to be there. They aren't. They're going there because Facebook can provide those things, and those things are meaningful. It's meaningful to me that as I grow older and my time is more and more committed to family and work, that I can connect with my best friends from my home town. That means something to me.
I'm at work so I can't write a lot more, but I am super happy to answer questions. I just hate these articles that assume that people are dumb, that social media is inherently vacuous, and that UX designers have some evil necronomicon of addiction recipes. It's cheap journalism.
Disclaimer: Everything here is my own opinion and not that of my employer in any way shape or form.
No offense intended, but given Google's "privacy for me but not for thee" secrecy and anti-leak stance, this kind of makes it hard to take some of your claims at face value. I imagine honesty could put you at risk of job loss or worse.
If documentation for implementing dark patterns existed at Google, I don't expect you'd be disclosing that information.
Full disclosure: I find Google pretty repulsive as a whole, so that could be clouding my thought process here.
Of the eight dark patterns you identify as such in your thesis, Facebook implements at least two, possibly three - I could swear I've seen reports of what you describe as "hellbroadcast", but I don't have a cite to hand. Your "impersonation" I've seen very recently reported here on HN at least, and what even is Facebook if not the ultimate example thus far of what you (correctly) term a "social pyramid scheme"?
In general, while I'm willing to assume good faith on your part, I do find myself moved to inquire further into how you compose, on the one hand, having produced detailed documentation and an inventory of examples on how to develop user experiences which are designed specifically to the purpose of maintaining user engagement in ways largely orthogonal to the inherently engaging quality (or lack thereof) of the underlying content - and, on the other hand, your argument here that platforms like Facebook, which are so broadly known to implement such patterns that to advance the claim merits no controversy whatsoever, do not achieve any benefit, or their users suffer any detriment, from such implementations.
One further point: I don't see the article here under discussion assuming that "people are dumb", or that "social media is inherently vacuous". Instead, what I see is a very reasonable question being raised around, on the one hand, the ethics of social media platforms like Facebook using these techniques to drive engagement, and, on the other hand, how people might protect themselves from being so manipulated - a subject which, while I haven't yet read it in detail, seems to be covered in your thesis as well, under the name of "manipulation literacy". I look forward to reading your exegesis of the topic!
Of the eight dark patterns you identify as such in your thesis, Facebook implements at least two, possibly three - I could swear I've seen reports of what you describe as "hellbroadcast", but I don't have a cite to hand. Your "impersonation" I've seen very recently reported here on HN at least, and what even is Facebook if not the ultimate example thus far of what you (correctly) term a "social pyramid scheme"?
Fair point. I didn't mean to imply Facebook as a paragon of goodness, but rather that social media as a whole does not inherently imply dark patterns. I think there are certainly things I'd like to see removed from products and I think those products would actually benefit from that in the long-run. The problem is measurement: it's much easier to employ a dark pattern, show some short-term metric that indicates "engagement" and get promoted or leave before it all drops off again. Playing the long game is much harder, career-wise.
I do find myself moved to inquire further into how you compose, on the one hand, having produced detailed documentation and an inventory of examples on how to develop user experiences which are designed specifically to the purpose of maintaining user engagement in ways largely orthogonal to the inherently engaging quality (or lack thereof) of the underlying content - and, on the other hand, your argument here that platforms like Facebook, which are so broadly known to implement such patterns that to advance the claim merits no controversy whatsoever, do not achieve any benefit, or their users suffer any detriment, from such implementations.
This is a good question.
I'll address it backwards: as to "merits no controversy", again I just meant that social media doesn't inherently imply dark patterns. I often feel like media presents the argument as such: "X does A, Y does B, all social media is bad by default." It's a sexy headline, but I don't think it really holds a lot of water with the majority of the 1B+ daily users of FB.
I do think a whollleee lot of research is needed into how people's social relationships are changing and whether their real happiness is going up or down, but I think of social media as part of the solution, not the cause. I would have grown a lot more distant from my social connections due to work/family as I grew older whether or not MySpace existed. I am not sure if at the time of writing the thesis that I actively avoided discussing users as "happy" or "unhappy" but I am glad I did. Motivation and happiness are not necessarily the same thing (Frodo is super motivated to get to Mount Doom, but he's not happy about it) but I think UX designers really do intend their users to be happier.
As to how I square the circle on writing a book with all these patterns, but then turn around and say companies don't have reference material like this... First off, I didn't contribute to it no-one read the book :) It probably sold 50 copies or so. When I wrote the thesis, gamification was beginning to cool off from it's peak bullshit, but there was still plenty of bullshit to go around. I chose the topic to sort of make the statement "if you are going to do this stuff, at least try and do it so it's actually trying to meet a users needs rather than just chucking dark patterns at them." I hoped that if the book became popular, we'd see less of the bad stuff and more thoughtful application of the good stuff so that apps would be more useful/meaningful to users.
When I was doing my research, I went barking up a great number of trees trying to find evidence of some sort of secret documentation of motivational patterns (it was helpful studying in Silicon Valley). I never found it. Doesn't mean it doesn't exist, but I never found it and I really don't think it does exist. I think designers go with their gut and what worked elsewhere, and they don&...
This does indeed help! I greatly appreciate your thoughtful response, and can only apologize for my inexcusable tardiness in following up further.
I'm not sure it needs secret teams of behavioral economists to produce dark patterns which drive motivation even to the detriment of the users they artificially engage. You can hypothesize and A/B test your way into such patterns as well, I think, and never quite realize what you're actually doing - for those of us who develop software on a professional basis, how many of us can maintain the perspective of an end user into the software we ourselves work on every day? Certainly I can't, and the next one of us I meet who can will be the first. Granted Facebook's engineers operate on a higher level than the norm, but in that regard? If anything, they're far more isolated from their users than the norm, too.
I'm sorry to hear your book sold poorly! I have to admit I've contributed to that trend, since I have no application for the information it contains. But I assume the book must differ markedly from your thesis, which appears much more concerned with factual documentation of the patterns it describes, and much more muted in its expression of opprobrium toward the dark patterns you cite there. I'm not sure I agree that the use of even those patterns you recommend is as innocuous as you suggest, but that's something on which we can reasonably disagree, I think.
Finally, I'd note that my concern with Facebook isn't so much its effect on its users, and on those who aren't its users but are surrounded by those who are, today - to be clear, I'm not at all sanguine about today's effects either, but what concerns me far more is that Facebook has almost without notice amassed an extent of power utterly unprecedented in human history, and it has done so in a fashion which permits that power to be exercised in total opacity - quite aside from the limited nature of options for curbing that exercise should it prove noxious, it's very hard even to know when and how that power is being used.
Until recently, Facebook has been satisfied simply to exist, and to grow. After the events of 2016, though, we've already seen initial gestures on the part of Zuckerberg et al toward a more active exercise of that power - and whether or not you agree with the direction in which it's been suggested that power be exercised, I should think the mere existence of that power, in the hands of people who can do what they damn well please with it and never need fear being gainsaid, should be worrisome in its own right. Perhaps it's being used today in a way that accords with your preferences. Will it be ever thus?
I do, though, take your point about the conflation of social media with Facebook, and I agree that it's understandable but not helpful that people equate the two. Facebook's model is, I think, an accident of history - when it came along, there were no good ways for people to both connect socially on the Internet and maintain ownership and control of their identities and information, and Facebook solved the first problem well enough that for a long time nobody much thought hard about the second. But there's no clear reason why both can't be simultaneously solved, and I know of at least one credible project currently underway which intends to do precisely that.
Re: dark patterns and being muted:
a) If it was too impassioned, it wouldn't have read very scientific.
b) From a purely self-interested standpoint, I honestly did not want to taint my ability to have a career. Expressing hard line views are for tenured professors, not PhD grads who are thinking they might go into industry :/
Thanks! And I suppose I can't really argue with that.
I mean, I could, but not with any conviction; on the one hand, I opted against college and found my early adulthood less circumscribed in some ways thanks to an absence of terrifyingly sizable debt - but on the other hand, I'm not about to try to pretend it's anything other than the clear partiality of almighty God, or if you wish a frankly implausible degree of good fortune, that's made it possible for me to build, on the strength of a high school diploma, even something vaguely resembling a software engineering career. Were it not for so many things going my way that sometimes it keeps me up at night just with the thought of how things would be if even one of them had broken differently, I'd be doing well to make $15 an hour in a job that prompted daily fantasies of suicide. So any criticism I might presume to offer, of the fashion in which you've achieved all that you have, would I think have to be pretty laughable on its face.
> makes it sound like users are forced to be there. They aren't. They're going there because Facebook can provide those things, and those things are meaningful.
No, not really. Drugs aren't inherently meaningful, they mask needs that aren't being met, hiding pain or sadness temporarily. They don't meet the underlying need.
That sounds like anti-drug propaganda. People with unmet needs can fall into drug abuse but drugs plain and simple have effects on the mind that can be profound. To assert that, in contrast to meaningless drugs, the majority of stuff on Facebook is meaningful is, well simplistic at best.
Unlike potato chips? What product is not designed to be as addictive as possible? One could argue they are being more manipulative but this is not unique or special to social media. Anyone remember subliminal messages in advertising?
I don't think capturing it as "addictive" is totally fair. If people don't get utility from these hooks, they'll leave. Growth hacking doesn't ensure long-term retention, product utility does.
"You know when you open Instagram or Twitter and it takes a few moments to load updates? That's no accident. Again, the expectation is part of what makes intermittent variable rewards so addictive. This is because, without that three-second delay, Instagram wouldn't feel variable. There's no sense of will I win? because you'd know instantly. So the delay isn't the app loading. It's the cogs spinning on the slot machine."
Can anyone confirm if this is true or not? Do social media apps insert `sleep()` calls on startup for gamification reasons? I've only seen that for things like save dialogs, but I also don't work in social media.
It could happen by accident. They insert a new feature into the app which increases load times a lot. It A/B tests better. Not because the feature itself is good, but just because increased load times do better.
We know these companies have people working to exploit psychological quirks/flaws to increase engagement. We also know that ethical concerns are somewhere below the bottom of the list of considerations when creating/implementing new "features".
Given the things we know, I'd be more surprised if this wasn't true.
I don't mean this exact case, I mean stuff like intentional slowing down of results. People don't trust results that come too quickly. We like to feel that the computer worked hard for it. E.g. Airline prices or TurboTax results.
One thing that helped me manage my addiction to Facebook was the realization that the number of notifications I get is nearly constant no matter how long the interval between using it is.
That is, if I am frequently active, I am reacting to the same number of notifications as if I am occasionally active.
The more you react to social media, the more it reacts back. It is a feedback loop.
The funny thing is that when I think about this, it always comes down to government regulations; when big corporations (whose interest is always to maximize profits at any cost) are allowed to get inside your pocket, your kitchen, and whatnot, and spend billions in learning how to influence our brain's reward systems without little limitations, I am not surprised they achieve so much success, after all, we are limited by our biology, we like it or not.
Same could be said of TV/Sports/Games/Any form of leisure
The only difference in social media is that we have actual data on how we're spending our data, vs. in the past when we didn't. To suggest that this is new is silly.
Perhaps, perhaps there's a point to be made about the scale of social media and how many people it impacts. But it's hard for me to believe, without any data, that it's tangibly harming us more than any other form of leisure.
109 comments
[ 3.0 ms ] story [ 221 ms ] threadReaching for addictive behavioral models is like reaching for management policies when the hard work and long-term investment into leadership is more sustainable.
To directly respond to your question, they can for example move to a facilitation model; engagement is encouraged through longer discourse, or through editorial staff whose job is to summarize viewpoints/trends/patterns that quickly bring newcomers up to speed, but they do not moderate.
Another possibility is an aspiration model that shows the small steps and decisions it takes to succeed, rather than orient towards the highlights reel view favored today. Say someone you admire and follow wakes up at 3 am to go work out: the social media connection time-shifts to your timezone, corrects for your distance to your gym, and encourages you via your smartwatch to do the same, in realtime. They eat exactly 100g of oatmeal for breakfast; the social media connection corrects for your body measurements (lean body mass, gender, height, etc.), and says you can eat 90g to follow your aspiration person.
Yet another possibility is community-wide goal setting and pursuit. Engagement through groups creating their own layouts, platform-based applications, etc. to both decide and pursue their goals.
There are more models to consider. I posit they are not pursued by FB because they decided they cannot monetize those models under their current ad-driven monetization model, which in turn is governed by the capitalist (though not necessarily free market) model they operate under. Relax some or all of those constraints, and the other kinds of engagement models might emerge?
Especially when you consider the success ... or more accurately lack thereof ... of attempts to deliberately counter the tendency.
I think this is fairly deep psychological wiring, and possibly informational dynamics, period.
This from someone who detests the practice and would much rather see alternatives.
Examining the simplification of choices to binary is a nice example, IMO, of how to make social media feel less "work":
> Following the introduction of Facebook's like button in 2009, YouTube moved to a binary like/dislike format in 2010. Instagram launched that same year and came ready-made with a Like function shaped as a heart. Twitter adopted this same heart-shaped system in 2015, while, in the years since, Silicon Valley has come up with a multitude of new ways to gamify our need for social validation.
I've never really liked FB's multiple emotions, though that is partly because of the interface design in which it just takes longer to register one of the other non-Like emotions. Netflix's move from stars to thumbs-up/thumbs-down felt controversial, but I'm pretty sure I've "thumbed" more things in the past month than I had made star ratings in the past year, simply because it was easier. And the data that Netflix provides me in terms of potential movie appeal -- a likelihood described as a percentage, versus a five-star scale -- feels more straightforward, and doesn't seem to significantly differ in accuracy from what I remember of what Netflix used to recommend for me. Roger Ebert's print reviews used a 4-star scale, but his Thumbs-Up/Down with Gene Siskel was still very popular and widely influential.
On the topic of social media engagement design, I would love to hear more about the recent Facebook feature that prepends my name before every notification. e.g. "Dan, you have 32 new notifications and 3 pokes today." versus "You have 32 new notifications and 3 pokes today." It's such an easy an obvious feature to implement, I wonder how long it was in the making? e.g. A/B testing, psychology studies, etc. Can't say it's really worked on me except to make notifications feel more of a nag but I'm likely an outlier in that I generally rarely visit FB.
The ethics chapter is awesome by the way...
For example, the article quotes the author of a different pop science book on the same subject; quotes a different source on ethics; and references intermittent variable rewards which were pioneered by BF Skinner almost 90 years ago.
Or that the user will scroll more and see more ads.
This technique is also called "good writing" in some fields. You see it in newspapers a lot.
http://websitecopywritingservices.com/blog/bucket-brigade-co...
In disagreement with the parent's snarkiness, I would argue there's a fine line between efficiently structuring written information to convey important news in a time sensitive manner and using bright red numbers to keep users swiping your app all day. Especially when it comes to messing with peoples self-worth and validation!
"You may have won $100! Watch these 5 advertisements to find out!" is gross but palatable to me.
"People might like you! Refresh the page and view another ad to find out!" is appalling.
Whereas "People might like you! Refresh the page and view another ad to find out!" cruelly exploits a desperate emotional vulnerability. And there is no end of it in sight.
That's why I now help to develop Lyra, an open, ethical conversation platform which respects language and attention. www.hellolyra.com
We're also a nonprofit with no marketing budget, so excuse the plug ;)
https://hellolyra.com/c/343
>1. No front page (for lurkers): They force users to join to be able to view content. The 90/9/1 rule is real, ESPECIALLY in the "community" space. 90% of users just want to lurk and that's just the nature of the internet. It's probably the same here on HN. 9% will participate in some ways, voting, etc. 1% will submit and comment, etc.
[1]: https://news.ycombinator.com/item?id=14411499
Is this a concern internally?
This makes me curious. How was the company culture while you were there? Did the company treat its employees any better?
My impression is that Lyra targets a niche of individuals that "enjoy conversing." I'm not sure if there's a nexus, a central target topic of conversation (like a sub-reddit would have) or a connective tissue such as real-life familiarity (like a friends and family social network would have) to hold your users together.
Can you elaborate on who your target audience is, and how they will be attracted to the site, or what will maintain their interest (while avoiding the trap of making their engagement addictive?)
Our target audience is those who want to converse online without the restraints and commitments of social media services which aim to addict rather than to support and convey language.
We don't aim to enable the discovery of interesting links - Reddit has this completely down. But it's no good for conversing in a focussed way with people you know.
Letters and email link people internationally in ways that are impossible on Facebook or Twitter. When you receive an email, you read it in a focussed space without distractions; and you're aware of the fact that it was personally sent to you as opposed to broadcast to a large channel. Narrowcast will always be a personal experience than coming across a message delivered by a mysterious algorithm.
Lyra does have a public news feed in which conversations are viewable to all users. If you add people to your groups and set the feed to view only these, it's the exact equivalent of Facebook.
What we don't do is offer particular spaces (hashtags or subfora) because these instantly offer a target for harrassment and abuse, and require moderation. Lyra's design makes it trivial for people and topics which are vulnerable to abuse, to control their reading and avoid it.
While the concept seems really interesting, what do you mean by "open" in this context? For example, are you able to export your data? (I wasn't able to find anything about that on the site.)
The app unlocks the site for a configurable amount of time after I've completed my tasks.
Also supports Hacker News, Reddit, YouTube, Twitter, and Product Hunt.
It has helped me a lot with a. Recovering lost productivity and b. Feeling less addicted to these sites.
If curious: Todobook https://chrome.google.com/webstore/detail/todobook/ihbejplhk...
https://en.wikipedia.org/wiki/Decompiler#Legality
Also, recently I realized that when I in nervous procrastination mode I also check GitHub too much to look at the stream of updates and "someone starred ..." stuff, I have looked for disabling the stream but could not find it... I guess I could try play with the Stylish plugin to make it disappear or sth.
Admittedly, sometimes I wish I could remove all "points" and "numbers" from the web, even on HackerNews. There is this quick brain sugar rush when you see these numbers rise, but just a minute later I feel empty, sad, and dirty for having this narcissistic automatic response.
This is one of the reasons (the other ones go around privacy and tracking) that I don't do Google Analytics nor any kind of access.log aggregation/analysis on my websites. Maybe this is a stupid rationalization, but I'd like to think that we value stats too much and that they distract us from doing better qualitative design and analysis...
Web news is generally like this, making it hard to see yesterday's stories today. That's one reason my preferred news source is https://en.wikipedia.org/wiki/Portal:Current_events
(We count 'day' using UTC so there's overlap with today depending on where you are.)
I recommend his TEDx talk[0], his essay on the topic[1] and finally, his Time Well Spent[2] project to get a firm grasp about his side of view.
(0) https://journal.thriveglobal.com/how-technology-hijacks-peop...
(1) https://www.youtube.com/watch?v=jT5rRh9AZf4
(2) http://www.timewellspent.io/
This seems to be happening not just on social media, but throughout the consumer technology space. The most obvious example is how video games skinner-box the customer, but I've also become cynical of how my industry (television software) is driven by consumer addiction. When I started working on television applications years ago, I saw it as an opportunity to help people experience art. Now I see people spending every spare moment glued to video, and I just feel like a drug dealer. :/
https://lh6.googleusercontent.com/-bpIIJYLwDps/TXzOnVIwaPI/A...
edit: Personally left Facebook a little while ago. It had reduced in my time to the odd few minutes scrolling threw the news feed. It just did nothing for me. I don't miss it.
I have never understood this complaint. Why don't you just unfollow those people? How many of them could you possibly know? This seems like a problem that can't exist for more than a week or two if it actually bothered you, since unfollowing someone when you see low quality posts like that will clean up your feed pretty rapidly.
(Note that unfollow is different from unfriend on Facebook)
It kind of feels like when I quit smoking...
The consumer doesn't receive any real, tangible benefit from the product. If they did, they would be willing to pay money for it. The only way social media, and much traditional media, can profit is by collecting attention and selling it to advertisers, so that's what they are optimized to do.
Let's see, keeping up with your friend? That used to be the telephone, for which people paid by the minute - an hour long call with your friend or family in another state might run $5, every time you wanted to talk to them. Or you could mail a letter. Slower, but still costs money for paper, envelopes and a stamp.
Wanted to share photos? First you had to buy a camera. Then you had to buy film. Then you had to pay to get the photos developed (extra for duplicates). These are clearly things that are (or at least were) worth money to people.
But a stream of random snippets from people you haven't spoken to in 20 years, asinine clickbait, and political rhetoric? Yeah, nobody wants to pay for that.
which I turned into a book: http://a.co/3uR0koj
I find these sort of articles annoying. I've not read the book referenced in the article, but I've seen some of Tristan Harris and I think he's a smart guy but stuff like the 60 Minutes interview plays into him looking like a hack who is far more interested in self-promotion than in the actual topic.
It just doesn't compute that "knowing what drives users" == "addiction manipulation". Yeah, there are slot machine-style techniques that you can use in those areas, but they don't work well or for long and users grow to resent them.
But the stuff that keeps getting brought up, like "people on Facebook are desperate for validation and connection" makes it sound like users are forced to be there. They aren't. They're going there because Facebook can provide those things, and those things are meaningful. It's meaningful to me that as I grow older and my time is more and more committed to family and work, that I can connect with my best friends from my home town. That means something to me.
I'm at work so I can't write a lot more, but I am super happy to answer questions. I just hate these articles that assume that people are dumb, that social media is inherently vacuous, and that UX designers have some evil necronomicon of addiction recipes. It's cheap journalism.
Disclaimer: Everything here is my own opinion and not that of my employer in any way shape or form.
If documentation for implementing dark patterns existed at Google, I don't expect you'd be disclosing that information.
Full disclosure: I find Google pretty repulsive as a whole, so that could be clouding my thought process here.
In general, while I'm willing to assume good faith on your part, I do find myself moved to inquire further into how you compose, on the one hand, having produced detailed documentation and an inventory of examples on how to develop user experiences which are designed specifically to the purpose of maintaining user engagement in ways largely orthogonal to the inherently engaging quality (or lack thereof) of the underlying content - and, on the other hand, your argument here that platforms like Facebook, which are so broadly known to implement such patterns that to advance the claim merits no controversy whatsoever, do not achieve any benefit, or their users suffer any detriment, from such implementations.
One further point: I don't see the article here under discussion assuming that "people are dumb", or that "social media is inherently vacuous". Instead, what I see is a very reasonable question being raised around, on the one hand, the ethics of social media platforms like Facebook using these techniques to drive engagement, and, on the other hand, how people might protect themselves from being so manipulated - a subject which, while I haven't yet read it in detail, seems to be covered in your thesis as well, under the name of "manipulation literacy". I look forward to reading your exegesis of the topic!
Fair point. I didn't mean to imply Facebook as a paragon of goodness, but rather that social media as a whole does not inherently imply dark patterns. I think there are certainly things I'd like to see removed from products and I think those products would actually benefit from that in the long-run. The problem is measurement: it's much easier to employ a dark pattern, show some short-term metric that indicates "engagement" and get promoted or leave before it all drops off again. Playing the long game is much harder, career-wise.
I do find myself moved to inquire further into how you compose, on the one hand, having produced detailed documentation and an inventory of examples on how to develop user experiences which are designed specifically to the purpose of maintaining user engagement in ways largely orthogonal to the inherently engaging quality (or lack thereof) of the underlying content - and, on the other hand, your argument here that platforms like Facebook, which are so broadly known to implement such patterns that to advance the claim merits no controversy whatsoever, do not achieve any benefit, or their users suffer any detriment, from such implementations.
This is a good question.
I'll address it backwards: as to "merits no controversy", again I just meant that social media doesn't inherently imply dark patterns. I often feel like media presents the argument as such: "X does A, Y does B, all social media is bad by default." It's a sexy headline, but I don't think it really holds a lot of water with the majority of the 1B+ daily users of FB.
I do think a whollleee lot of research is needed into how people's social relationships are changing and whether their real happiness is going up or down, but I think of social media as part of the solution, not the cause. I would have grown a lot more distant from my social connections due to work/family as I grew older whether or not MySpace existed. I am not sure if at the time of writing the thesis that I actively avoided discussing users as "happy" or "unhappy" but I am glad I did. Motivation and happiness are not necessarily the same thing (Frodo is super motivated to get to Mount Doom, but he's not happy about it) but I think UX designers really do intend their users to be happier.
As to how I square the circle on writing a book with all these patterns, but then turn around and say companies don't have reference material like this... First off, I didn't contribute to it no-one read the book :) It probably sold 50 copies or so. When I wrote the thesis, gamification was beginning to cool off from it's peak bullshit, but there was still plenty of bullshit to go around. I chose the topic to sort of make the statement "if you are going to do this stuff, at least try and do it so it's actually trying to meet a users needs rather than just chucking dark patterns at them." I hoped that if the book became popular, we'd see less of the bad stuff and more thoughtful application of the good stuff so that apps would be more useful/meaningful to users.
When I was doing my research, I went barking up a great number of trees trying to find evidence of some sort of secret documentation of motivational patterns (it was helpful studying in Silicon Valley). I never found it. Doesn't mean it doesn't exist, but I never found it and I really don't think it does exist. I think designers go with their gut and what worked elsewhere, and they don&...
I'm not sure it needs secret teams of behavioral economists to produce dark patterns which drive motivation even to the detriment of the users they artificially engage. You can hypothesize and A/B test your way into such patterns as well, I think, and never quite realize what you're actually doing - for those of us who develop software on a professional basis, how many of us can maintain the perspective of an end user into the software we ourselves work on every day? Certainly I can't, and the next one of us I meet who can will be the first. Granted Facebook's engineers operate on a higher level than the norm, but in that regard? If anything, they're far more isolated from their users than the norm, too.
I'm sorry to hear your book sold poorly! I have to admit I've contributed to that trend, since I have no application for the information it contains. But I assume the book must differ markedly from your thesis, which appears much more concerned with factual documentation of the patterns it describes, and much more muted in its expression of opprobrium toward the dark patterns you cite there. I'm not sure I agree that the use of even those patterns you recommend is as innocuous as you suggest, but that's something on which we can reasonably disagree, I think.
Finally, I'd note that my concern with Facebook isn't so much its effect on its users, and on those who aren't its users but are surrounded by those who are, today - to be clear, I'm not at all sanguine about today's effects either, but what concerns me far more is that Facebook has almost without notice amassed an extent of power utterly unprecedented in human history, and it has done so in a fashion which permits that power to be exercised in total opacity - quite aside from the limited nature of options for curbing that exercise should it prove noxious, it's very hard even to know when and how that power is being used.
Until recently, Facebook has been satisfied simply to exist, and to grow. After the events of 2016, though, we've already seen initial gestures on the part of Zuckerberg et al toward a more active exercise of that power - and whether or not you agree with the direction in which it's been suggested that power be exercised, I should think the mere existence of that power, in the hands of people who can do what they damn well please with it and never need fear being gainsaid, should be worrisome in its own right. Perhaps it's being used today in a way that accords with your preferences. Will it be ever thus?
I do, though, take your point about the conflation of social media with Facebook, and I agree that it's understandable but not helpful that people equate the two. Facebook's model is, I think, an accident of history - when it came along, there were no good ways for people to both connect socially on the Internet and maintain ownership and control of their identities and information, and Facebook solved the first problem well enough that for a long time nobody much thought hard about the second. But there's no clear reason why both can't be simultaneously solved, and I know of at least one credible project currently underway which intends to do precisely that.
Re: dark patterns and being muted: a) If it was too impassioned, it wouldn't have read very scientific.
b) From a purely self-interested standpoint, I honestly did not want to taint my ability to have a career. Expressing hard line views are for tenured professors, not PhD grads who are thinking they might go into industry :/
I mean, I could, but not with any conviction; on the one hand, I opted against college and found my early adulthood less circumscribed in some ways thanks to an absence of terrifyingly sizable debt - but on the other hand, I'm not about to try to pretend it's anything other than the clear partiality of almighty God, or if you wish a frankly implausible degree of good fortune, that's made it possible for me to build, on the strength of a high school diploma, even something vaguely resembling a software engineering career. Were it not for so many things going my way that sometimes it keeps me up at night just with the thought of how things would be if even one of them had broken differently, I'd be doing well to make $15 an hour in a job that prompted daily fantasies of suicide. So any criticism I might presume to offer, of the fashion in which you've achieved all that you have, would I think have to be pretty laughable on its face.
Yes. Connection. Is that's we've seen a huge decline(40%!) in empathy among college student in 2000-2010 [1] ?
[1]https://www.psychologytoday.com/blog/born-love/201005/shocke...
Same arguments could be made by a drug dealer.
The one that makes apps more addictive is https://useDopamine.com
The one for making thing less addictive is http://youjustneedspace.com
I don't think capturing it as "addictive" is totally fair. If people don't get utility from these hooks, they'll leave. Growth hacking doesn't ensure long-term retention, product utility does.
Can anyone confirm if this is true or not? Do social media apps insert `sleep()` calls on startup for gamification reasons? I've only seen that for things like save dialogs, but I also don't work in social media.
We know these companies have people working to exploit psychological quirks/flaws to increase engagement. We also know that ethical concerns are somewhere below the bottom of the list of considerations when creating/implementing new "features".
Given the things we know, I'd be more surprised if this wasn't true.
This sort of thing is well-known, so doubting this particular case is not a well-founded doubt.
https://www.theatlantic.com/technology/archive/2017/02/why-s...
Habitual is the word they are looking for.
2. Turn off push notifications or turn on Do Not Disturb (you can whitelist favorites, all contacts depending on your OS)
3. Be intentional when you pick up your phone
The only difference in social media is that we have actual data on how we're spending our data, vs. in the past when we didn't. To suggest that this is new is silly.
Perhaps, perhaps there's a point to be made about the scale of social media and how many people it impacts. But it's hard for me to believe, without any data, that it's tangibly harming us more than any other form of leisure.