I believe their secret sauce is mainly that they don't manipulate the feed for other factors than if it fits.
The Facebook feed used to work like that before 2013 something. You could clearly see the difference when they started optimizing for ads or what ever metrics they used. Posts could not go "viral" anymore in the way it used to. Zuckerberg capped the vitality so that companies had to pay I guess.
They also don't just optimize for consumption, but also for new users. Don't have concrete proof, but it seems they help new users get more views to give them that taste of "15 minutes of fame" as early as possible.
Yeah I tend to agree regarding the 15 minutes of fame for new users. It's doesn't even need to go insanely viral.
How many people have created and published videos on YouTube and got maybe a few tens or a couple hundred views over years.
TikTok new users gets a couple THOUSAND views and the user is like "people like my content I MUST make more!!" And now they're hooked trying to get the same high view count as before.
I would assume (as I have never used the platform) it's also a way for TikTok to grab you financial information "you earned x$ from you first video, please supply you financial information and a scan of your ID for payment.
I don't know much about the partner program but if I change the wording to this:
"You _could have_ earned x$ from your first video. Good job! You're on your way to the Partner Program please supply your financial information and a scan of your ID for future payment."
Since you seem to have experience. Did you notice 15 minutes of fame on your first uploaded video? How many views did your first video get? What was the topic and soundtrack of your first video? I'd love to dig a little deeper.
Similarly, the various dating apps I have used clearly front-loaded the matches with the objectively attractive women in the area. This has been especially hilarious to me as a female friend will often start the process for me (search store, download, sign up, start searching) and more than once they have gotten this skewed version of what's out there, thinking that every nearby woman looks like a movie star.
You don't have to look that far for proof, there are plenty of people who made videos complaining about how they don't get good views anymore, they aren't pushed out anymore. Search "why isn't my tiktok account being pushed anymore" and you'll find similar outcry on various forums. You see it by scrolling through their profile enough, they go from 300k views and 50k likes to 1k views and 100 likes. Massive drops, where they haven't reached "self-sustaining" status yet like a Selena Gomez account.
Also as another potential source of proof, tiktok is really effective at pushing alt pages to people's FYP. Which is interesting because their old account will stagnate, but their alts will get pushed quite aggressively despite the content being very similar. So obviously it's not a content issue, tiktok is evaluating them equally and thinking you'll like it, but their old account(s) don't show up as much anymore.
Tiktok isn't as overtly aggressive with this method though, it seems like there's a cooldown period of sorts where it might start pushing you out again if you remain committed to the platform. Still, it's severe enough for people to make note of it and feel seriously demoralized by it.
I think a key part of their new user optimization is putting very little weight on your previous performance. So if you make a great video as a user with 0 followers, you still have similar odds to a 10M follower user to making it big on the FYP.
There might, of course, still be additional new-user weight. But I think this is at least a part of it.
I don't, no matter how much they ask. The most I've acquiesced to is a small amount of YouTube time per day. It gets harder the older they get. My daughter can now just use her school laptop (while at school, at least) to watch as much YouTube as she can find time for. I'm not real happy about it, but part of raising kids is learning to accept the steady transfer of their autonomy over to them. I just keep gently harping on how low quality this kind of content is, what the motivations behind it are, etc.
There's dogshit content on YouTube (Mr Beast, moronic reaction and gaming videos, alt-right propaganda), but there's also more interesting content than you can watch in several lifetimes (documentaries, speeches and lectures, recorded courses from top universities on everything from physics to music, etc).
It's about curating this yourself for your child (as opposed to giving them unrestricted access), and then slowly giving them control and hope they have good taste :)
It’s funny you lump in MrBeast with the other two. I’ll admit his content is 0 value, but in terms of entertainment, it’s far higher than anything else on the platform. I’d be far more upset at my kids for watching a reaction YouTuber or any influencer, than I would mrbeast.
He's a downright clever guy who knows exactly what he's doing to maximize entertainment value, and, by proxy, clicks and therefore revenue. He was on the Lex Fridman podcast a few weeks back, pretty good listening.
I absolutely detest those videos, and I have to say I don't understand the appeal at all. It's a combination of of unhinged consumerism and that uncanny fake excitement which is extremely off-putting to me.
Can't agree more! Facebook is useless to me today and replaces meaningful to me stuff with ads and other things I don't care about and never engage with, i.e. there's for sure other weights and the feed is not organic!
I've tested in the past and counted streaks of 30+ consecutive posts in my Facebook feed without any posts from friends, groups I'm in, or pages I've followed. Just "recommendations" and ads for miles.
I did a test as well. Unlike most people, I've set up relatives very well. Yet not a single post by a relative and some of those have high engagement! All kinds of Instagram Reels and ads and a bunch post from people with high a high absolute engagement, but historically a very low one from me.
So, this obviously is a poor filter, which does not keep me engaged and which proves that Facebook is NOT a social network as to be social, it needs to act as such, and I never hear from my closest people over there anymore!
How hard it is to see that a post is from my mother-in-law and multiply its weight by 100, or one from my sister and multiply by 1,000?
Another thing that Facebook never puts any effort into is photos of children under 13! They had a feature called "scrapbooks" where you can create profiles of your children with the promise of those turning into profiles after kids turn 13. Well, they've abandoned this feature and people kept tagging children as their parents! Not sure how this served their facial recognition, but it inconvenienced us parents!
It seems ad prevalence is directly inversely related to the market in that in a down market, we get flooded with ads to temporarily increase the co's cash flow with more ads, but people this time around are claiming it has completely shot the various recommendation engines.
The most impressive Facebook's feature was hands down Graph Search [1] which was released around this time - 2013. Basically it was an interface to the entire social graph!
Eventually they had to pull those features out. Privacy concerns which lead to people sharing less data? Not sure.
Even now a lot of people get surprised when I put their full name into search box and show them what photos are available via simple search - they completely forget or don't know even that when they are tagged their photos are searchable by their name. I mean now most people I know don't share much anymore, facebook is not that cool anymore.
Probably because open access to an entire community’s / nation’s social graph is a huge security risk. I mean if the NSA having access to all your phone call metadata isn’t already bad enough, why would you want any and all malicious state actors, criminal organizations, terrorists, or run of the mill scammers to have access to that type of information too?
I was behind a woman today on an escalator. I wasn't shoulder-surfing, she held her phone up so high I couldn't help but see it. She was on Tik-Tok. She watched like twenty videos in one minute. One second, click. One second, click. What good this does anybody at all, anywhere, is beyond me. It was all trash.
Agreed. She didn't necessarily decide to click through 20 small clips of lazy cheerleading. She was probably hoping to find a good hit but is satisfied with useless crap because it's better than being present.
Raises an interesting question: is something only trash if there are better options available? I'd say no, personally. Something can just be bad in isolation, it doesn't have to be worse than something else to qualify.
But at the very least, maybe it was just trash compared to the experience of not watching a short video on your phone while riding on an escalator. Trash compared to just looking around. Would staring at your shoes, or the head of person in front of you, or the environment surrounding you have been a better experience than those short videos.
As a recovered TikTok addict, I can supply that it was the jokes for me. There was a lot of content that got delivered to me that hit my sense of humor just right. I’m sure you could track increased oxytocin in active users just like with narcotics.
Throat cancer, too. Friend of mine smoked cigars regularly for much of his life. Once they diagnosed the cancer in his upper throat, he lived about 10 months.
i'll pile on: my dad had jaw cancer. cut a few chunks of his jaw out, basically lost all of the molars on the bottom left. probably going to have to take more out before too long.
also turned into a few skin cancer issues in there.
How is that any different from other social media feeds? Users familiar with a content platform will skim through content faster. That has been the case since the TV era.
It's not just the viewers either. I guess they need to be there for someone to bother creating content, but on one short stroll through London I saw multiple "content creators" filming themselves. That wasn't something that one often saw when I was younger, now the whole city is crawling in people who are making videos.
I also noticed the quality of the filters. They really make ordinary people look quite a lot better than they do in real life.
> What good this does anybody at all, anywhere, is beyond me. It was all trash.
It's TV 2.0, but caters to ADHD types or people with borderline ADHD. If you use The Internet in any extreme way, you already have an ADHD brain. It's the only logical way to handle information overload and deal with the vastness of the web IMHO. ADHD as a term is normalized because of The Internet. It's not a scary thing to have anymore. It's the new normal.
All that brilliance and knowledge, used to get vulnerable, suggestible teenagers addicted to mind-numbing, anxiety-inducing media. The shame of software engineers is that we have been used to create 1000s of these inhuman products.
The problem is not just with the addiction, and it doesn't affect only teenagers.
These systems show paid content mixed with user-generated content, promoting harmful advertisements, political propaganda, or any other ideology anyone with enough resources and desire to influence large segments of society is able to take advantage of.
Let's not forget the Cambridge Analytica scandal, and not doubt that there are _many_ more such companies operating without public knowledge. This is all done legally, and within the terms of service.
This is even without mentioning the rampant data siphoning and shady multi-billion dollar data markets that exist in the background, that generate most of their revenue. After all, all this technology and optimization is serving this end goal.
These platforms are not just harmful to individuals, but to society as a whole. I hope one day governments catch up to the harms they're doing, and regulate them just as they did for Big Tobacco and many other truly evil industries.
>I hope one day governments catch up to the harms they're doing, and regulate them just as they did for Big Tobacco and many other truly evil industries.
But how does that happen when our government officials (at least in North America) seem technologically illiterate, and rely on Silicon Valley for their re-elections and donations (therefore becoming more susceptible to lobbying)?
My feed is full of mostly educational content, lawyers and other professionals telling stories, language lessons, old computers and tech, and old commercials.
My question is why teenagers gravitate to horrendous content en masse.
Young girls have feeds full of eating-disorder content, and young boys have feeds mixed with misogyny like Tate.
Not all teenagers watch either, but these recommendations are a disease plaguing the platform.
They don't gravitate towards it. The algorithm inches them towards that. It starts with exercise videos and healthy eating, which is fine, but eventually goes towards eating-disorder content. Similar for boys.
Here's an analogy to understand ad-driven social media apps:
Imagine a world where there are thousands of companies out there that produce these strange little unpleasant pellets of food that they want people to eat. They don't want people to buy them—the company is happy to send out pellets for free because they make money when the pellets get eaten. (Let's not worry about how that could make economic sense. Maybe the pellets are made of sequestered carbon and the company is selling carbon credits.)
So all these companies want to get millions of people eating their pellets, but people don't want to. They don't taste very good and they've got, like, better shit to do with their lives.
But people do like eating other food. And usually they have to pay for it, which kind of sucks.
An opportunity exists here, and "social eating" companies pop up. These companies will ship you food for free, and you can eat as much as you want. What joy! The only catch is they've mixed some of these not-very-tasty pellets in with the food they send you. The social eating companies get paid by the pellet companies, which is how they're able to make and send you food for free.
Here is the interesting question: What kind of food do the social eating companies make? You might think that since they want to get as many people eating their food mixed-with-pellets as possible, they would make the best food they can. And, indeed, they do want to make food that is scrumptious, compelling, and mouth-watering. But what they don't want it to be is satisfying. Because once you're sated, you stop eating. That's the last thing they want because the more pellets you choke down, the more they get paid.
So what they make is junk food like chips and cookies. Each serving is a single tiny delicious bite, but as soon as it goes down, you're even hungrier than before. Its high in anticipation and desire, but low is satisfaction and satiety. You crave it, but once you have it, you don't actually feel any better. If anything, the craving is even more intense.
Now apply that metaphor to information. The most junk-food-like content is the stuff that triggers anxiety and anticipation if you don't watch it: fear of missing out, question-inducing "You won't believe...", alarming "You're doing ___ wrong...", etc. The content doesn't make you feel good if you do watch it, it makes you feel bad if you don't. Because that way, when you do watch it, the nagging anxiety goes away a bit, but you still don't feel "done" and still want more.
that which inflames or causes controversy gets sent around or reposted.
clicks == money.
as does "sponsored content" aka "post this to everyone who meets demographic groups X and Y". just so happens that some of those people want folks angry about some topic.
I guess that makes sense. Just following the user's preference gets only positive content, which is great, but carefully steering them to something darker is even better.
Agreed, but I don't want to excuse anybody. I think every one of those engineers signed up to do it in exchange for money. They used their talents, they weren't forced into it.
I wonder what may happen in the future. I'm sure neither Purdue Pharmaceuticals nor the Sackler family expected to be seen as criminal in what they were doing (opioid epidemic). But when the outcomes were shown to be a result of their actions, they were convicted— or at least forced into settlement.
This is reductionist. Why does anyone want you to click ads? Because they want you to see what they shared with the world. The best minds are trying to get people to pay attention to those who wish to pay for that.
A video game programmer might sat.. "all of the best minds are working on silly games" Why are so many bright minds wasting trying to get us into space when we have problems that need to be solved here.
The best minds solving problems is only part of any solution anyways. You need the best salesmen, best leader and right moment to bring in change that mat make lives better.
> Why does anyone want you to click ads? Because they want you to see what they shared with the world. The best minds are trying to get people to pay attention to those who wish to pay for that.
Framing advertising as "people trying to share beautiful things with you" is some hell of a newspeak.
I hate to break it to you, but if you think TikTok has "grass roots content creators" vs Facebook having influencers, I have a bridge, as well as a TikTok content house to sell you.
But they do seem to be optimizing more explicitly for "user enjoyment" than other platforms, perhaps realizing that this is what keeps and attracts users and their precipitation eyeball-minutes in the long term. Youtube also (not always) does a pretty good job at this, while Instagram can be rough.
Another possibility is that TikTok's short-form content feed allows for a greater volume of feed to fit into the same timeframe. Twitter was popular because it condensed blogging into microblogging; TikTok's success is pretty easy to delineate from that logic. Even if you don't have an algorithm promoting enjoyable/affirming content, you're just more likely to encounter that content when you watch more videos.
Whether that behavior is healthy or not... I'll leave that up to people more educated than I. My guess is that trading the user's attention span for short-term engagement isn't worth it, though.
I'm with you, and I'm of the opinion that the current TikTok craze has more to do with its brand than its algo. Facebook is old and lame to kids. Even Insta and Snap don't have as much brand cache as TikTok right now because TikTok is the newest major player with the most buzz. However, it's the nature of all hype cycles for high-flying, shiny new things to eventually fall back to Earth:
It's way easier to find "smaller" content creators on TikTok than Meta's apps. Youtube has been improving a bit in that regard but my recommendations still have too many videos I've already watched. More than half of the accounts I follow on TikTok have <25000 followers, many with average views per video under 10000, with a diverse topic range including: woodworking, software dev, comedy sketches, book reviews, history trivia, exercise tips, urban design, etc. Those who complain about seeing only unintelligent garbage are unwittingly revealing themselves to be the target audience for such content.
The main reason TikTok has awesome recommendations is that all the other big players (Facebook, Youtube, etc) realised back in 2017 that training machine learning models to persuade people to spend more time on their platform was unreasonably effective.
Ie. tell ML to trick someone to spend 14 hours watching youtube every day, and for some small percentage of users, it will actually succeed!
For those people, it's as addictive as drugs. They spent all day on youtube rather than going to work, going to school, caring for their kids, eating or even sleeping! Can you imagine the size of lawsuits that would be heading youtubes way when those people realise they've effectively been enslaved by an algorithm??
Leadership of the big companies put an end to that, instead trying to focus on other metrics, and trying to get more users to each spend some time on the platform.
I know someone, mid 30's, jobless, who spends 12+ hours a day on TikTok. They don't even post to it - just watch hours and hours of mindless videos about mental health issues.
Maybe not 12+ hours a day, but a good chunk of time and this is my house mate. They watch more Tik Toks about ADHD than they do actually trying to live with ADHD.
You probably engaged positively with some of those videos. The feed tends to give you more of what you engage with.
Mine is full of cats, food, attractive women, and also ADHD content. None of those are remotely surprising to me, except that there's not more ADHD content (though I suppose the feed implicitly is).
It's not "bad" content. It's content you don't want, but which your behavior nonetheless suggests you do. If you're not seeing things you like, search for a hashtag you're interested in.
No, it's objectively bad content. Only due to the way they track and understand user behavior suggests that the content is correct to show to the user. That's not the users fault when they get shown crap stuff
Which is sad, but sans TikTok, would they just be on Reddit or Facebook or Instagram or YouTube or Netflix or Hulu or Disney+ or Paramount+ or HN for 12+ hours instead?
Well Facebook failed because almost no one i knew using it in 2017 still is. It’s like the normal reaction to an addictive thing is to get captured by it for a while, realize the downsides, and drop it after a time. Some people are bad at this and stay addicted but most people escape and go on with their lives. Optimizing for short term addiction sabotages your product.
My facebook feed is a ghost town. Almost zero people i know posting, a couple of straggling groups that remained relevant, and just bs ad driven nonsense.
People will pick their own poison to fill their free time, in the past it was reading books then it became watching TV all day im sure most millennials are using apps to fill their free time.
I used to fill my free time with world of warcraft spending more time in Azeroth than i did with friends or at school or at work. But i came to the realisation it was just escapism from a boring country/region and a waste of my time. But this is also made me realise how lucky i was to be born in such a boring country/region.
I like those charts showing they get an AUC of 0.8 or so, my own content-based recommender gets 0.72 or so on a typical day with a very simple model, I think it could get 0.75 with mainly data-oriented tweaks. I was thinking a better model could get me into the high 70's but the fuzzy nature of the recommender problem means I am never going to get into the 90's.
I would like to see more about how they are formulating the problem, I know there is work lately in "sequential recommendation" that is focused on generating a sequence rather than scoring content items, I'd like to learn more about that.
Some tasks are hard, others are easy. I get in the low 60’s trying to predict if a headline will get a lot of votes on HN. Of course this is going to be bad because the same article can get submitted several times, hit it big once and not hit it big the other times. The content of the article matters too but the model can only make a wild-assed guess. I get in the high 70’s trying to predict if a post with a lot of hits gets a lot of comments, it is a cleaner problem. I get about 90% trying to predict if a post winds up ‘dead’, I found up HN is cleaning up a lot of spam posts when they go in but I know it is terrible overfit because it says an arXiv paper which has no chance of being flagged has a 30% chance of flagging.
For some other like “is this an astrophysics or materials science paper?”, getting 0.95 or better was straightforward in 2005.
If you had to pick one number for judging classifiers that is the one.
"Accuracy" is a really bad one because it is not meaningful w/o careful thought about the problem. For instance if you have a diagnostic system for a disease that has a .01 prevalence, you get 99% accuracy if you say nobody has a disease.
Practically a person might want to operate a classifier with different thresholds depending on how bothered they are by false positives vs false negatives, the ROC shows what choices are have available with a particular and the AUC of the ROC gives you one number that characterizes the quality of the classifier overall. A system with an AUC under say 0.55 hasn't really learned anything at all, at 0.6 it is showing signs of life, looking at those curves I'd say Tik Tok is beating me solidly. In this case it is limited by the problem being fuzzy (something I thumbs up today might get a thumbs down tomorrow), something up in the 0.95 level could be attained for something like "is this article about the outcome of a sports game?"
To be fair tiktok helped me find a lot content (series, movies, etc) and artists (specially standup comedians) that it would take a much longer time to cross-paths with. So I would see a short of that, and then looked around to watch the whole thing.
But in the 1 year I've interacted with the app I've already seen the effects of the content creators learning how to game the algorithm and get presented with some contents that were made in a format only to get engagement, similar to a clickbait article.
The result? It has spoiled the whole experience for me and I uninstalled the app.
And this is how those platforms slowly crumble to their own weight as the crowdsourcing becomes a business for creators and, little by little, content becomes more and more superficial, fake, commercial and uninteresting for the audience, who migrates to another app and everything starts again...
In these articles I never see reference to the 'objective algorithm' aka what the business processes at TikTok are optimizing for. I think that's the real secret sauce.
TikTok is optimizing for user generated content that connects deeply with their users. Thats what they reward money which is largely derived from TikTok's "gift economy", and recommendations and therefore views are largely based on the gifts. This is in stark contrast with something like YouTube, which is optimizing for 'time spent on YouTube' which drives their creators to make longer clickbait type videos, regardless of how dissatisfying they are.
The recommendation algorithm described here could be tuned to serve any 'objective algorithm', like time spent, profit to Bytedance, or whatever- but because of TikTok's choices it is making a really fun an enjoyable experience for users.
Why would people spend time watching dissatisfying videos?
> content that connects deeply
I'd say "parasocially".
But how does the app measure that? It makes decisions based on a few seconds of watch time.
It could optimize for "show videos that trigger donations", but that's kind of trivial.
There might not be a simple topline objective everything is optimized for. More likely it's a linear combination of all kinds of dynamic business needs.
As a person who occasionally consumes TikTok videos, I feel like it taps into the same thing that the original StumbleUpon did; show the user fairly quickly a number of related things and give them the option to either consume more of that same thing or find something else that's related but also new. Same goes for Pinterest. They all share that core element of quick consumption of related things in common.
Pretty click-baity title. This article is a good "ELI5" summary for non-ML people of some of the techniques & infra described in the arxiv paper, but it has nothing to do with TikTok "secret sauce".
What's addicting about it? Admittedly it kept me entertained the first day or two that I used it, but from that point forward it never wavered from the content that entertained me that first day and soon it became really boring to see the same type of thing over and over, leaving no remaining appeal.
Because everyone talks about how great the recommendation engine is, and me wanting to find great content, I went back to it after a while to see if they fixed the problem. But no, still the same content that bored me away from the app the first time.
I had the same experience. I’ve given TikTok three different tries over the same number of years and it didn’t connect with me at all.
I’ve spent several hours using it and there seems to be very little of interest to me (a 53 year old Canadian living in Texas). It did show me a video of a guy doing some great welding (stack of dimes!) and I watched the whole thing. The algorithm decided that I must want to see every welding video ever uploaded and I really don’t. One or two is enough.
I concluded that TikTok isn’t for my demographic, maybe by design.
I found it took some curating to get the algorithm to what you're interested in - like videos you enjoy, and aggressively mark videos you don't (hold and select "not interested"). In my experience, by the end of day, the recommendations were more on target.
Maybe TikTok appeals more to teenagers who have an underdeveloped sense of self. I remember at that age conflating popularity with accuracy and authority. Now when I watch TikTok it's just an exercise in frustration that nothing is cited and it's all about presentation over substance. And that's when the algorithm isn't showing me strange videos in a desperate attempt to keep my attention, a la spiderman/elsa YouTube mashups of yore.
I believe you but you should know you are in the minority here. Part of why TikTok has taken over the social media space is because of how addicting it is compared to the older players.
I can completely understand how it keeps your attention longer than looking at photos of your neighbour's cat like Facebook gives you. It's just the lack of variety that makes addiction hard to fathom. Do others not get bored of seeing the same thing over and over and over?
To avoid seeing the same thing over and over, the most direct form of feedback that works for me is when that content comes up, quickly swipe past it. If you keep swiping, it'll start throwing more random stuff at you to see what sticks. I'm also always tweaking the list of people I follow so that if FYP gets stale, I can just look at their content for a bit.
I have heard that before, and tried it briefly, but it didn't accomplish much in the short time I tried. I can believe if I put a lot more time and effort into it that might change, but the last thing I need is another job. The whole value proposition of trying TikTok was the promise that it would do that work for me.
They're testing features to let you "reset" your algorithm to improve content diversity. But even without that kids spend over 80 minutes a day on TikTok.
Do you have the stats for that? I recall reading something earlier this year than Insta is still king for teens/young people. I can't find the source though.
For me, the moment I see two videos in a row of the same format and with the same music under it I get a sense of existential dread and close the app. Something stops feeling human and that's why I deleted my tiktok and might delete my Instagram next
This blog post is unlikely to contain TikTok's secret sauce, or even anything that is actually used by TikTok.
ByteDance runs TikTok, but they also sell recommendations as a service.
The paper that this blog post is based on explicitly talks about that service in the abstract, and never even mentions TikTok.
Now, if I were to advertise a service that contains some of the ingredients of my other, very popular product, I would surely mention that fact ... but this is not the case here.
Clearing out old and stale IDs from the model is interesting. It makes sense for a platform serving a high volume of viral / short lived content. I can't imagine the same trick working at a streaming service.
Pro-tip: On Instagram, interact with ads that you don't mind. That way the algo picks those or similar ads which makes a more pleasant experience. For example, now I mostly get ads of nice pictures of watches.
this is a wall of text citing a single paper which itself is entirely unverifiable.
regardless, if this described how drug dealers engineer the distribution of their product to optimise addiction, we would not be celebrating it as an engineering marvel.
The Facebook feed used to work like that before 2013 something. You could clearly see the difference when they started optimizing for ads or what ever metrics they used. Posts could not go "viral" anymore in the way it used to. Zuckerberg capped the vitality so that companies had to pay I guess.