I worked at Facebook for some time, and did a bunch of data work. We had this culture of building something, looking at the results between the experiment groups, and then choosing the statistically more successful one -- i.e. a newsfeed algorithm that had better engagement.
This sounds great at first, and certainly is straightforward if you want a promotion. But behind the scenes some of us had this thought that our observations only amounted to short-term gains. Although we had small long-term experiment holdout groups, the truth is they were rarely reviewed because it was unsexy.
My current thinking is that features like the echo chamber effects from Facebook's algorithms, Snapchat's snap streaks, and clickbait like this, all serve to optimize short-term engagement. Yeah, I want to watch that sexy new show or keep my streak going or have my opinions validated. But there's a diminishing return on clickbait, hollow articles isn't there? I can only fill up so much time with garbage like that before I'm bored. I can only like so many posts before I feel like they're all the same. And once my snap streak is broken I hate snapping.
The data/engineering/product loops at tech companies favor boosting short-term metrics; The employees are incentivized to do so and this is what they measure, so this is what they build. That's why we end up with features like this. That's why Snapchat fell off. That's why Facebook fell off. And that's why Netflix feels increasingly stale (despite there being a lot of quality content if you dig).
Humans are instinctively drawn to a lot of things that are really bad for us in significant amounts, because we are designed to fit into an environment where those things are scarce. Sugar, fat, inactivity, "interesting news", outrage. These all steal our attention because that response helped our ancestors survive and pick out these rare but important treats. But in typical human fashion we've now crafted a world that gives us these things all the time, and it's making us sick and miserable as a result.
"But we're just giving the people what they want" some might say. Well, depends on the definition of "want". In a taste test, junk food would win over broccoli for me, hands down - but it's broccoli that ends up on my table more often than not, because it's what makes me happier and healthier in the long run. It wasn't until my thirties before I realized just how sluggish I got after eating junk food. But if Facebook was running our diets, their algorithm would long since have "optimized" its way to junk food for all of us.
Sometimes the worst thing you can do is give people exactly what they ask for. Being healthy in a world like the one we've created requires much, much more restraint and self-discipline than it used to. "The algorithm" is basically the digital incarnation of the little devil on our shoulder whispering that we should treat ourselves.
> But if Facebook was running our diets, their algorithm would long since have "optimized" its way to junk food for all of us.
I really, really like this comparison.
If we fed our bodies in the same way we let tech company algorithms feed our minds, we’d have three Big Macs for breakfast washed down with a shot of whiskey and a couple lines of cocaine
When I read that, I kind of thought "hasn't that already happened"? The most enduring symbol of American capitalism is caffeinated sugar water. Our athletics competitions are sponsored by fast food conpanies.
We live in an obesogenic environment. Yeah, people with reserves of health, wealth and wisdom can buck the trends by applying time and energy to fighting against the tide but if the average person is falling for this stuff, the battle's already lost as the insanity becomes normal.
I feel like on the nutritional side there's been a (not total, but significant) reversal over the last decade. Some of that gets eaten up by marketing (food that isn't any healthier gets packaged in muted shades of green and brown to make it appear more grounded and wholesome), but some of it is real (I no longer know a single person who drinks soda on a regular basis).
Maybe in another decade or two we'll see a similar shift in the digital space.
I don't think that's the whole story. Even national brands like McDonalds have subtly changed their packaging to feel less artificial and more "earthy", for lack of a better term, even if the product itself hasn't changed. There's an awareness of a cultural shift.
I would guess that processed/high-sugar foods today are where cigarettes were in the 90s. Everybody knows and accepts how bad they are, many people have changed their habits accordingly, many people haven't.
Digital media on the other hand is more like where cigarettes were in the 70s. Most people know roughly that they're bad for you, but few do anything about it.
Yeah people are free to choose but to go full circle many of us let ourselves be manipulated by advertising[0] and dark patterns[1] to make unhealthy choices. That's why USA is so obese.
When I look at countries like Japan I don't see a limit in food choices, but I see a lot more healthy people.
Back in UK history there were paid rabbit hunters, who'd roam a farmer's land hunting rabbits. They ate a lot of rabbit. It turns out that rabbit meat is not nutritionally complete and some of them died.
I have another related theory that goes: Over time, everything converges into mediocrity. By everything I mean products sold to us, entertainment, media, the Internet itself. Look at the car designs of the 1970s and now. Look at cinematography then and now. As soon as marketing steps in and start telling otherwise innovative companies or artists how to sell, things begin slowly converging into mediocrity.
The semi-formal proof is that because the humanity as a whole is mediocre almost by definition, hence if you want to expand your markets (the ice-cream saleman's problem) you'd grab the middle of the bell curve first: it's where the bigger part of the market is!
I don't know if it makes sense, just something I've been thinking about a lot lately.
I was driving to work (a foreign experience) and Joe Walsh was singing about a Maserati. And it made me thing “man, if you grew up in middle America, hearing about a Maserati, you might not even know what it was, let alone that it existed”. Now you can hit Google and see millions of pictures, join the fan club, watch hours of videos and reviews, and become an arm chair Maserati expert. Which is cool, but if sort of takes away the mystique behind things. You wouldn’t even know how it was spelled unless you had the album itself and the artist included lyrics.
Similarly with guitar, it would have been almost impossible to figure out what type of pedal someone was using to get “that sound” without interacting with other humans in person. And I think that served as a filter for people who cared and also led to interesting introductions and local experts (music shops and store owners, etc).
I think that’s what we lose. And while I think converging to average makes sense, I think peoples tastes have shifted to quality in more areas, because they can quickly google and see what the best looks like. And most people will use debt to get there.
I can identify with the guitar example. I started learning to play in high school, right before internet access became common for average folks in the US (I knew some kids who had CompuServe at home and I was aware of some BBSes, but nothing like today).
I remember going to the music store where they had a room with sound dampening and a giant pedal board hooked up to an amp. Some friends and I would head over after school, guitars in gig bags, and plug in to "try out the pedals". I'm sure the staff loved us since we couldn't really afford to buy them ever. But we got to hear what they all did.
Other times I'd think I invented something, only to realize it had already been done. I had this little headphone-amp that ran off a 9V battery and plugged into the 1/4" jack on your guitar. It was great to play loud (through headphones) and not annoy my family.
But one time, I had some old broken headphones plugged into it and found that I could wedge the driver in the corner of my mouth, play distorted guitar through it, and shape the sounds with my mouth. Do it in front of a mic and whoa! How cool is this??
Later on I learned talkboxes were already a thing. Do you feel like I do, indeed.
One of the cognitive hacks I suggest running on yourself is when you create something, then later discover that it is in fact something already well known, rather than being disappointed that you didn't invent something new, take it as validation that you were on the right track instead.
“man, if you grew up in middle America, hearing about a Maserati, you might not even know what it was, let alone that it existed”
I assure you that even a rural kid like me knew what a Maserati was. We had movies, TV, and car magazines. People's assumptions about information delivery pre-internet are horribly warped.
Fair - I think my point is more along the lines of seeing one and getting a detailed look at the features. That was reserved for car shows, but now there are hundreds of YouTube videos that go into a lot of depth.
And immediacy - before you’d have to wait if you didn’t have car and driver handy or the movie with the Maserati in it wasn’t rewound. Now it’s 30 seconds and you’ve got more information about them than you could look at in a lifetime.
Growing up, I didn’t know what certain words in songs were (because I couldn’t hear them clearly or just hadn’t heard the word). It meant finding album notes, looking for the lyrics (if they were there), then trying to use context clues to put it together.
> I think peoples tastes have shifted to quality in more areas, because they can quickly google and see what the best looks like.
I've noticed that I do a lot more min/maxing. I'm always optimizing my purchases for the best available at the lowest cost. Which means hundreds of hours of research. Research which often makes no real difference in the end.
> I think peoples tastes have shifted to quality in more areas,
I'm genuinely curious if there's any data or research to support this. Because in my view the tastes remain more or less the same, just the information noise around us is now bigger.
I agree with the sentiment that designing a product for mainstream appeal can make it seem bland, but I don't think that goes all the way to "Over time, everything converges into mediocrity."
There are so many things today that are better than in decades past. Computers are unbelievably faster. Electric guitars are much more consistently high quality. Cars don't break down as often.
Maybe the problem is that the word "mediocrity" is too big. It encompasses both quality (which I think is mostly improving) and uniqueness vs blandness.
Is the quality really improving though? People complain about how washing machines and other appliances last shorter these days, so that you buy/upgrade more often.
As for cars, a lot of the times they look so similar that you can't tell the brand at a distance. They were certainly more distinguishable decades ago. Today it's also the high-end mobile phones - they all look like the iPhone, which isn't bad per se, but if you can't innovate (or don't want to) that's mediocrity.
I agree with you, particularly when it comes to information consumption. On-demand services are bad enough; cleverly optimized services that push content are even worse.
But in some ways, if you hang out in the right circles, being healthy is a lot easier today. Alcohol consumption seems to be down, people exercise a lot more, being out of shape in your 30s is not seen as 'the norm'. So it is probably easier psychologically to go for a jog now when parks are full of joggers than to do it in 1950s when people would think a 30-something person out for a jog is a bit strange.
In this way, healthy fads can actually be very positive. I find it very likely that in the near future a culture will emerge that tries to minimize the use of technology and information hygiene becomes a part of 'normal' life.
> Sometimes the worst thing you can do is give people exactly what they ask for.
Alan Moore said this in the context of poetry, but with the caveat that everybody is sometimes part of "the" audience, and that every fully fledged human adult should sometimes be "the" artist, I think he has a hugely important point:
> It’s not the job of the artist to give the audience what the audience wants. If the audience knew what they needed, then they wouldn’t be the audience. They would be the artists. It is the job of artists to give the audience what they need.
Isn't that more because of what they are optimizing for than the fact that they are optimizing? To continue the food analogy engagement is blood sugar levels and they are seeing how high they can crank it. If they optimized for "happier and healthier in the long run" then they would probably loose short term engagement spikes, but have fewer people quitting for health reasons. Basically they could be broccoli if they wanted to be but they choose to be soda.
Well, yes, they could certainly try. But "happier and healthier in the long run" is much harder for them to measure than "has responded to instant gratification". It's much harder to A/B test effectively at longer time scales and make the ML model reinforce the right things.
But also, "healthier" might involve creating less engagement "content" for them to feed to others, lowering the network effects of the whole platform. And if they successfully go that route, they open a flank to anyone willing to just keep spiking blood sugar as much as possible (TikTok?). There's a reason McDonalds are still selling Big Macs, and it's not that they don't know what healthy nutrition looks like.
I work in SRE where a pretty common expression is “what gets measured gets fixed”. I used to take that as at least mildly inspirational, and to mean that more and better monitoring leads to more things being fixed. And to some extent that’s true.
In recent years though I’ve come to see the downsides of that mantra as outweighing the good of it. Because some things are either extraordinarily difficult or expensive to measure, or because understanding what the measurement is demonstrating is beyond the intellectual reach or experience of many people. By the latter sentiment I mean, it’s not enough to just show a number or a graph, it has to be interpreted, and for some things that interpretation is very challenging if you’re not a (or the) expert in that system.
As a result, it’s more like “easy to measure and understand things get fixed, everything else gets ignored”. It disdains or glosses over the idea that maybe a person or team’s subjective opinion about what’s important to fix carries any weight at all, because if it was really so important, surely they’d be able to demonstrate that in a form that someone (possibly willfully) ignorant of the system can understand.
I see the same forces at work here, in marketing and a/b testing. The simple to understand metrics are what are optimized, while the more complicated ones get ignored or drowned out. The longer term benefits are hard to measure, and more importantly, hard to understand and interpret.
While it is harder, it isn't impossible. They have done QOL studies to find out what kind of posts make people feel depressed and what not.
Totally agree with that second paragraph though. Leaving openings for competitors is how you get killed by competitors. Not sure where the line between compelling but not evilly so is.
The problem here is that if there was one single company that wanted to exit the standard approach and create less engaging content they would be more likely to fail.
The companies that are creating content which optimises to the most engaging content will drown out the one "healthy" one.
Well with food there are readily available numbers that approximate the long-term-effect of food (e.g. saturated fat, preservatives, vitamins, glycemic index). Does a company really have a way to quantify the emotional impact on its users of a code change? And if it did, would they optimize for that or focus on profit margin?
all-broccoli diet is no good either. so it’s the optimization itself, yes
for a balanced diet, we have to make our own choices, based on our particular, local situation, and not rely on any generalised algorithm. which is very unpleasant. i wish there could be a Facebook telling me what i actually should know! alas, it's always going to be on me to choose
> The data/engineering/product loops at tech companies favor boosting short-term metrics
I think this hits the nail on the head. Leadership is hard, and executives can be many layers removed from the frontline. It seems that A/B tests and family have proven a really effective vehicle to communicate across leadership levels what has happened, why it happened, and the impact it had on the business. I'm definitely guilty as charged of having participated in this "pyramid scheme".
Teams get a tool to structure work and determine success/fail. Managers get a tool to detail exact impact to leadership on how they're solving business problems. Executives get a tool showing how teams are being deployed on thoughtful projects with an iterative and rigorous scientific approach that helps increase confidence around deploying capital. The board gets some slides during the quarterly meeting and maybe might even see some charts showing growth and less cash burn.
Eventually over longer periods like a year, executives will notice different trends than what the short-term numbers are showing and will have to make real hard decisions about how to steer the ship.
How might a streaming service be designed to increase user happiness and wellbeing?
From a positive design perspective, could certain streaming features give users a greater sense of purpose or goal accomplishment (e.g., mastering a genre) or help users meaningfully connect to others (e.g., seeing what friends have recommended lately)?
While measuring the effects of a feature on wellbeing outcomes is substantially harder than measuring the effects of a feature on engagement (time spent), it is worth measuring/optimizing in the long run. The hypothesis is that capitalist enterprises that make their users feel happier and more fulfilled will, in the end, be more successful than enterprises that merely provide empty satisfaction. If not, the future could be very bleak.
> How might a streaming service be designed to increase user happiness and wellbeing?
Yesterday I worked longer, because we had a critical bug at an unfortunate moment. At some point my brain stopped working and I couldn't make progress. So I stopped work for the day, solved some solitaire, mindlessly played some chess, and watched a couple episodes of anime on Netflix. After that my brain unblocked itself and the world stopped spinning around me. I wouldn't be able to read a book in this state, or anything requiring any thought or much energy[1]. Doing completely mindless things is sometimes beneficial for you.
[1]: You might laugh, but you can absolutely play chess on autopilot without any thought and still win with people.
I don’t think anyone is saying that mindlessness should be banned or completely removed from a platform. But that it should not be pushed so heavily that it is edging out any kind of deeper engagement, or just non engagement.
All these services are competing for human attention, which there is only a finite amount. I think it is pretty obvious that this has resulted in more mindlessness, where apps are trying to fill every sliver of open attention with the most easily digestible little piece of data (a 10 second meme, a couple of comments, a 30 sec friend snap). Combined with apps and services trying to take your attention from other things through strategic notifications, this is the opposite of long term wellbeing.
Attention and focus for long periods of time is a skill. Delayed gratification is a skill. Both of those are needed for long term growth, developing other skills, and doing activities that are fulfilling but not immediately gratifying. Being surrounded by things that constantly and perniciously attempt to carve up your attention and provide instant gratification is detrimental to those skills.
> designed to increase user happiness and wellbeing?
I have complicated feelings about the roles of business in our lives. I'm not entirely certain that's a metric I'm concerned with them optimizing on, nor do I believe we can even remotely quantifiably capture it in any meaningful with with human-computer interactions.
Complicated feelings are warranted. But is it better for massive companies to ignore human wellbeing?
We know that algorithmically optimized digital services can negatively effect wellbeing. We also know that management responds to metrics. I'm proposing that wellbeing metrics (namely, the self-reported effects of services on aspects of wellbeing) should be periodically gathered through surveys. Surely this takes effort, but if we only optimize what is easy to measure (time spent), this will continue to produce negative unintended consequences.
How do we even begin to define wellbeing? How do you measure it, and do so at scale with statistical confidence? How do you finance it? How do you moderate and govern it? Who does it benefit besides the industries that spring up to support it? I don't think "wellbeing" is concrete enough an idea like business metrics to meaningful act upon in private organizations. Sort of a "road to hell paved in good intentions" concern.
Here's how the OECD does it at a country-level. A couple decades ago this probably seemed too hard. Then, it turns out it's not that hard to operationalize wellbeing and measure it at scale with statistical confidence. It isn't perfect — the science advances — and that's good.
Companies already invest heavily in the assessment and optimization of employee wellbeing. This is a major industry in HR now, especially during covid. Is it really such a far stretch that a Facebook or Netflix should regularly gather metrics from a subset of customers to understand how their products affect user wellbeing?
After reading up on it, there's nothing about this that builds confidence that it's anything more an indicator of wellbeing than the "MyPlate" USDA guidelines are for dietary requirements.
And this is evidence somehow, other than being another research paper in a sea of many? I'm sure there's none of those at the FDA. I mean, the abstract ends with:
"Results suggest that there is considerable disagreement regarding how to properly understand and measure well-being."
The existence of research does not make something definitive, but I'm always happy to be proven wrong.
Even basic physical phenomena aren't definitive. That's not how science works. We shouldn't expect a singular metric of wellbeing to rely on for all time. Existing wellbeing measures and models are useful enough to inform practical decision-making and design. What we admittedly lack are measures sensitive enough to indicate when small design improvements are successful. More research needed.
> businesses are built to generate profit at the customers' expense.
Absolutely agree, which is why I'm bearish on the roles of companies. Companies are made up of people, and significant change comes slowly to people over very long periods of time. Look how long things woman's suffrage, unemployment benefits, and civil rights took. The U.S. is still fighting over single-payer healthcare while 66% of all bankruptcies are tied to medical bills, and we can't even agree on whether the Covid vaccine is a good thing.
I don't think I want random groups of people joined in coordinated capitalism to increase their profits at my expense, while constantly moving between those groups, also being responsible for my wellbeing. I want them to sell me their shit so I can get back to figuring out the rest of this mess.
The point is that they will support user wellbeing if it makes them more money. If half of society is saying "don't use Facebook, it's bad for you" that's not ideal from a business perspective. It's why McDonald's offers salad.
> How might a streaming service be designed to increase user happiness and wellbeing?
The rational part of my brain would like
- Suggestions for difficult content (this tends to be things you either love or hate, which is why Netflix does not suggest it, preferring safe bets);
- Making it impossible to binge on one, or a small set, of things, instead forcing diversification;
- Encourage spending short amounts of time -- e.g. by automatically breaking up movies into episodes, not autoplaying the next episode, disincentivising stupid cliffhangers, etc.
I realise many of these things are actively user hostile and a service implementing them would be dumped in favour of the alternatives faster than they have time to rollback the commit. Including by myself, when I just want a low-effort way to pass the time with my wife until it's socially acceptable to go to bed.
And that's sort of, I guess, the point. In order to increase user well being, a service like this probably has to encourage the user to spend less time with it. But that's not a business model that generates any money.
The last two suggestions are interesting to me because it's exactly the opposite of what I think I need for my long-term well-being. Over the past few years it feels like watching too many clips and reading too many short articles has caused my attention span to drastically decrease, to the point where I rarely have the patience for a normal length movie. Breaking movies into episodes or forcing me to watch more different types of content without finishing any of them would make this even worse.
I think a better way to limit mindless engagement would be the ability to pre-set and lock in which content I'm going to watch for the day. So maybe I lock in that I want to watch a movie or 3 episodes of show X, and then Netflix shuts down for the day afterwards. Maybe it could even warn that episode 3 is the first part of a two-parter and suggest I stick to two episodes for the day.
> I think a better way to limit mindless engagement would be the ability to pre-set and lock in which content I'm going to watch for the day. So maybe I lock in that I want to watch a movie or 3 episodes of show X, and then Netflix shuts down for the day afterwards. Maybe it could even warn that episode 3 is the first part of a two-parter and suggest I stick to two episodes for the day.
This is the right approach (giving the user tools that increase their agency) and the equivalent of the Digital Wellbeing features in Android.
I don’t think it has to be actively hostile. I think being almost neutral would go a long way.
-No autoplay and bringing you back to the main menu at the end of content would give your brain a chance to actively decide if you want to continue or do something else.
-No notifications or pushes to get you back on the service. So many services try very hard to take up space in your mind while you aren’t using them, in order to get you back on the platform.
-Suggestions allow for diversification as well as deeper understanding. This might be suggesting documentaries related to fictional subjects you are watching, critically acclaimed content in your preferred genres, or expert analysis and contextualization of content that allows for better understanding of cinematography or cultural influences on a piece of media.
-Low effort and binge content is shuffled out of immediate sight and flashy attention grabbing ways of displaying content in general is stopped.
Now you are more of a passive participant, where you keep getting notifications until you open the app, then are bombarded with easily digestible, low effort content that actively plays itself, and discourages anything that requires delayed gratification or long term focus. These things would shift the way you interact to an active participant, where you decide to engage the app without prompt, and look for something you want to engage with, or don’t and easily leave to do something else.
This really only works if every other app and service isn’t constantly battling to fill every sliver of open attention with the most easily digestible little piece of data (a 10 second meme, a couple of comments, a 30 sec friend snap) and to drag your attention back from other things through strategic notifications, gamification, and feelings of social obligation (did you wish kqr happy birthday? click here to write on his wall).
Attention and focus for long periods of time is a skill. Delayed gratification is a skill. Both of those are needed for long term growth, developing new skills, maintaining other skills, and doing activities that are fulfilling but not immediately gratifying. Being surrounded by things that constantly and perniciously attempt to carve up your attention and provide instant gratification is detrimental to those skills. I want my digital environment to help improve those skills, or at least not be actively harmful to them. In the same way we have spent the last 15 years and petabytes of data, optimizing algorithms to increase engagement by any means, we could use that data and AI to start optimizing for sleep quality, mental health, goal attainment, financial stability, physical health, minimization of insecurity, etc. If Amazon can predict when women are pregnant before they know, I’m sure we could optimize for these.
> In order to increase user well being, a service like this probably has to encourage the user to spend less time with it. But that's not a business model that generates any money.
That is really the fatal crux of a digital environment funded largely through ads. The user will always be the product and will be manipulated to benefit the customer, regardless of the effect on the user.
The answer is at the very bottom of the article and maybe Netflix is already doing it. It mentioned a producer was told his show was renewed not based on the "how many people viewed for at least two minutes" metric that gets reported to ratings publishers, but based on how many people actually watched the entire season.
It isn't this trivial for all media, but I think "watched the entire thing" is a very obvious metric to optimize for when recommending and promoting films and series.
The only thing worse than engineers trying to use algorithims to make us eat junk food is them using them to try and make us healthy. This is because at least the junk food part no one will be able to seriously defend is morally right. The best way to tyranize others is to claim to be working for their own good.
Quality often escapes measurement. The naive overreliance on finding a number to optimize is an enemy of quality, and is all over the place, and can be boiled down to people being driven by somebody’s need to make a nice powerpoint slide.
I generally agree. Finding numbers to measure quality almost never helps composers or cooks, but it almost always helps organizations. Why? Quality oddly lends itself to quantification. Rotten Tomatoes and Amazon stars don't always work, but i find they work shockingly well. I'm not going to naively defend the benefits of quality measurement, but i do advocate for the importance of continuously improving quality measurement systems. And key to this, imo, is making sure we take the effort to measure the values we actually want to enhance.
Totally. But what's the metric for good? If it is the average time spent watching or most likely to be chosen out of a slate of options, then we get the current problem.
Early in his term, Trump flirted with the idea of changing quarterly earnings reports to be biannual. I wonder how much that would have shook up corporate culture.
Kind of a pity his term went in such... surreal directions, because I think he probably could have had some interesting ideas in areas like corporate governance. He had a mix of insider experience and a populist platform.
They don't seem to have fallen off, are still the most widely used platforms in the world with billions of users in total. It is popular to think that A/B testing leaves you in local maxima, but what about alternatives? The only thing I've seen work is something like Pixar's brain trust - you do testing but you also rely on the good judgement of a small group of people with a strong sense of vision. Judging good judgement is pretty hard though, and you won't know if they're just full of it or winging it unless you try and trust. And so everyone does A/B testing which to be honest sounds much better than your average PM making decisions that are pulled out of their ass.
> It is popular to think that A/B testing leaves you in local maxima, but what about alternatives?
The alternative involves creativity to create a vision to achieve. A/B testing are pretty good tools to evaluate differing opinions on some vision, as the only deciding tool for what is worth to keep or not it's pretty short-sighted.
Is it really that much better? I lived through the rise of A/B testing, professionally I've been responsible for implementing platforms for experimentation in at least 5 different companies and over the years I saw it transition from a tool to a crutch. Everything now is "data-based" which means incessant A/B testing, it still depends on your average PM deciding the parameters of the experiment, leading to the current state of affairs: a bunch of uninspired features that are solely judged on the data produced, no soul, no vision behind most of it.
And it feels exactly like that, I can feel how tech products just became that, an incessant stream of features being tested, implemented or tossed.
I'd much rather have data-supported intuition than this statistical machine in motion.
The problem is not only that it could be a local Maxima. It might also measure the wrong metric all together.
Is the platform widely used and the business successful because content makes users happy and improves their lives?
Or is engagement high, because users have become accustomed to whatever a company is feeding them and look for something shallow to fill up their time.
If you only measure engagement as metric, without any deeper reflection and judgment, then heroin is a perfect product in terms of A/B testing.
This is the consequence of lack of vision. What is Netflix's long term goal? Ideally, it would be to fund and broadcast fantastic cinema. Yet, I really don't get the impression that contributing to the art form of film is very high up on the metrics chart.
I'm of the opposite opinion. They're executing their vision perfectly.
Netflix is very aggressively pursuing the bottom line, and they're not doing it differently from a drug dealer. It's not random that the term "Netflix binge" exists.
Netflix interface is designed to shove content at the users' face at all costs, and in a very disrespectful way. Here there are some disgusting dark patterns:
- it took them a long time to give the user the option to remove movies from the watching now list
- there's no way to remove already watched movies from the panel; their trailer is even sometimes displayed in the top frame
- they randomize the global order of the horizontal lists, so that one needs every time to hunt the user list (watching now/my list) through all the panel
For a short time, I even experienced a bug where movies with (my) negative rating were not greyed out. I can imagine the engineering thinking "LOW PRIORITY!".
On my (fictional) "drug-dealing practices" rating, Netflix is high on top with Amazon.
So they are taking steps which are ostensibly geared towards this drug dealer vision. But is it working? I can only look at myself and the people around me. There was a time when we would go on Netflix binges and there probably was something drug-like about it. But now? I go on Netflix, I scroll around for a bit, can't find anything I would like to watch, and I close Netflix again. The only reason I'm still paying for it is that I'm sharing my account with other people and it would be a hassle to cancel the subscription. And it's not just me, everyone I talk with is reporting the same thing.
I’m in total agreement. Opening the Netflix app just started making me frustrated trying to find something to watch, so I just don’t bother anymore. I will only open it now if somebody has recommended something to me, which happens rarely these days. We’ve only kept the subscription going as our child likes watching some of the shows on Netflix kids.
They used to have all their content come from other studios, and they got to see how well it had done in the outside world at the time they were making their deals.
Now they're producing their own content, many of their previous background-noise or binge staples have moved elsewhere, and their productions are overwhelmingly lowest-common-denominator clickbait. There's still some quality stuff, but it's much harder to find than it used to be. If you're looking for an already-well-known quality show you can search (e.g. if you want to watch The Office a few years ago on Netflix). If you're looking for something new that's going to appeal to you, it's a lot harder to find in their sea of junk. Reviews can get you so far, but Netflix no longer then has any edge on alternative platforms which the reviewers are also talking about.
So the premise of "sooner or later pumping out junk food is going to turn off viewers" seems to click with your cohort (mine as well).
Maybe that's really Netflix's business model these days: nobody watches anything, but it's a hassle to cancel the account, so nobody does it, so lets continue frustration our users, maybe they will stop visiting us completely while never really leaving us.
Let's look at the bright side though, we're going to end up with a lot of nice thumbnails.
High use means you need large quantities of high quality content. Low use means you can use a small amount of high quality content as a "loss leader" then back fill with cheap garbage no one is going to watch anyway.
It depends on the viewer's attitude. I think there is a spectrum between the extremes "I want to watch something interesting" and "I want to watch something".
Considering that the commonly mentioned stats on TV watching show that a lot of people is watching (or maybe more accurately, keeping on) TV for hours every day, the latter extreme should be common.
In this sense, bombarding the audience with junk, or at least, pushing very hard, is a strategy that, I think, is productive, in the same sense as drug dealing strategies are.
I'm towards the former extreme, however, I'm certainly not immune from their strategy. I still feel some need to "chill" and therefore I somewhat tolerate their strategy (although, I also use it for learning languages, so there's another motivation). The irritation is building up, though.
> But now? I go on Netflix, I scroll around for a bit, can't find anything I would like to watch, and I close Netflix again.
I feel like I'm in a minority these days. My Netflix list is longer than I can ever hope to finish. I add stuff on before I finish watching current stuff.
100% agree. I was awaiting for Netflix thread to pop-up to ask (hopefully someone who works there): Why there is no way to tell Netflix that I already watched the show and I'm not planning to rewatch it again any time soon or hide the ones I tried and decided not watch (or may be watched before)? No matter how many times you show me them I'm not going to watch...
You can still have nice features around it: separate list at the end of scroll with my old movies/shows (may be with number of times I watched it?). Once new seasons released send them all the way to to the top etc.
And I will be able to discover something new. At the moment it is cluttered with things I don't want to watch so lately I was not happy with Netflix at all.
Add one more to the list - and this one drives me crazy to no end - on my Chromecast with GTV the wretched thing starts autoplaying even if I just want to read the synopsis. This, despite turning off all autoplay options in my profile. What is the friggin rush?! Why won't you let me browse/read distraction free?!
Look out for this annoying autoplay thing in every other app soon. I do some work in the same space and everyone* thinks if Netflix is doing it it must be a great idea.
The sad part is that I am probably in some minority and hence no one will care to fix scenarios for me. The only time I now watch Netflix is when I want to watch something specifically and I search where can I watch it and it turns out it exists on Netflix (of course, never recommended to me). I wonder how much long term will Netflix be impacted by the fact that their Originals are of way lower quality than HBO originals. I never want to watch Netflix Originals because whenever I tried it was horrible and HBO always has amazing quality. I'm afraid it's just cheaper for them to focus on the group of people who you can shove the shit down the throat to keep them happy as it's a way maintainable audience.
Unfortunately, the new HBO content is the same as Netflix. After HBO was sold to ATT a couple years, the executives that made HBO HBO were all shown the door and now I believe their goal is quantity over quality.
Apple TV+ feels like it could be the new HBO in terms of curation, but we will see if they keep up the quality ratio.
My simple method of not wasting time on all this nonsense content coming out is to just wait for a bunch of people in my various networks to mention something about the show/movie. If multiple of them like something enough to mention it, then I probably will, so I just search it out after that.
A lack of understanding their own business model? They are a gym, they need to provide enough service to get you to sign up and keep paying, while minimizing your actual use.
Has netflix ever committed to this vision? I would be hell bored if Netflix is filled with Oscar-ish films.
One thing to note here is, the tech community is inherently elitist, in a way many envision tech as a tool to doctrine its audience, telling them what is good or bad. I don't find this mindset, either helpful nor necessary.
Recommendation engine is the new soda machine, choices are offered, and you pick what delights you. Ofc you can choose water, but it is not the machine's fault some would like real coke.
As another former Facebook employee, I mostly agree with this.
One nitpick is that although you may have not looked at the holdouts often, somebody definitely did. Different teams use their holdouts differently, but leadership probably looked at your holdouts at least a few times per quarter. And now with recent infra and changes, it's very likely that data scientists or product managers somewhere in your org are responsible for explaining results to higher ups at least quarterly.
All that is to say, the cause isn't that Facebook (and Netflix) aren't thinking about or monitoring things in the long term. It's that they are measuring the wrong things, because it's very hard to measure things like "this is clickbait".
Yes, Facebook is now almost useless. The only reason to be there is to keep track of friends and family, but it has gotten to the point where if a family member has a big life event, I'm hearing about it through traditional in-person or on-voice-call word of mouth _faster_ than through Facebook, because Facebook keeps deciding for me what it wants me to see.
Netflix has kept trying to make it harder for me to find what I want for the last few years, and I use it less and less, and now pay for fewer accounts than in the past. They are in danger of losing customers, with their annoying mouseover/autoplay/clickbait interface.
Not my experience. A contractor recently told me almost all of his business were driven by his company's FB page. I'm also aware of a few small businesses who do most of their business via FB connections. FB is well and alive in the mainstream, even if it skews toward more older group of people.
Yes, I'd love to see a more robust approach to novelty decay rates (to make up a metric) in A/B testing, especially around engagement.
You'd think that would be of interest to the business to know that X "successful" intervention had a typical average lifetime of Y before reverting to the mean, for example.
Then again, there's the idea that competitive advantage for social networks is just finding enough novel interventions before the novelty of the platform itself is exhausted, and FB has no short supply of other platforms to milk in that regard.
Netflix has somehow conditioned me to expect disappointment in movies with interesting looking cover art, which I now actively ignore, and don't even click through for descriptions anymore.
"Huh, that looks interesting. Not falling for that old trick again! What's that half-off-screen ambiguous cover on the next row down?"
I think you can maintain the feedback loop without degrading the user experience if you actually ask them what they want to be shown.
The problem lies in that "engagement" definition : if the user clicked on a thumb, then they must have been attracted to it. In reality, there is considerably more explanation for that click : They could be looking for a specific title, mistakenly thought _this_ actor was playing in the movie, clicked on it to reset the search, wondered if this was a show or a film, missclicked, just picked the least annoying in an ocean of shitty propositions... An algorithm alone can never find its way for better suggestions if it lacks _intent_. The outcome of this lack of data and the lack of data scientists' imagination, will always be stuck in a local minima.
So, what solution do you have to gather this precious intent and act on it ? Well, you start small on predictable things : "what would you recommend to Agatha if they liked _Expendables_ ?" "Is _RED_ a good addition to the list if I'm looking for Action/Comedy films ?"
And later "Are you looking for Action/Comedy films ? here is some proposals"
It works the same for different approaches, for picking thumbnails, or promoting movies. Just ask user for what's motivate them. If Netflix had asked me last week, I would have told them their catalog is getting worse and worse so we prefer using disney+ instead.
The idea is that these kinds of sources of error will be noise, since they should be random with respect to experimental assignment in any kind of A/B test. In a small sample you could imagine chance imbalances, but given that I'm experimentally assigning someone to see either a thumbnail of John Krasinski or Steve Carell, it's unclear why the "I misclicked" scenario is going to have a higher rate in one rather than the other. And experiments at large companies like Netflix are almost infinitely powered.
I would think a bigger threat to this approach would be being stuck in a local maxima.
i do not understand snapchat and their business model. they started out as a sexting app that deletes the photos and that is where it remained for a long time. i remember news of how "deleted snapchat photos can be undeleted" tutorials and i tried to investigate for fun but i could not understand the UI back then and the tutorials didnt work so i dropped it.
Apparently its pretty big these days in the kids and they have "streaks" as you mention. i hear multiple hundred thousand streaks and other things but to what end?
how is snapchat justifying the storage, bandwidth and processing of data? i mean facebook has ads which they earn, same for twitter but what about snapchat? do they have a sustainable revenue stream because no where i heard snapchat had a paid option.
reddit was user funded for a long time with daily goals and all that so that was not an issue mostly but what snapchat done to address that.
snapchat is all the rage in my city, kids who are now on online classes send snaps because i hear about it. everyone knows how to make that twistface pout or whatever but why?
genuinely curious about it
> i do not understand snapchat and their business model.
Ads[1]. Initially it was just promotional lenses/filters and required a high touch process and substantial media spend commitment. But looking at that site, they now have a full on self-service ad platform with minimal spend commitment, helper software to make lenses/filters, and have added several other ad formats beyond that one.
So they've essentially monetized the same way as every other "free" platform has – capitalize on all the eyeballs they have to siphon off a cut of marketing dollars for themselves.
You've heard of streaks (which were a very early feature), and you've probably heard of Stories too, right? Video ads appear in between Stories video rolls.
It seems even worse than that. During my stints in social media, there would be ‘research projects’ that were simply fishing expeditions for meaningful data. Something like: which user demographics have a correlation to behavior x. And then you’d just search and search and search and hope you found something. If you didn’t find something, then you’d be encouraged by a team level manager to adjust or tweak some approach to take advantage of volatility to force the correlation. Maybe you shift the sample start date, maybe you skip a normalization dimension.
The directors would ask about things like normalization and if things were measured over a ncertain time period, but if you had some excuse or answer prepared, it was enough to get a nod and some recognition for your presentation. Heck, sometimes your team manager would do the presentation on your behalf to lay claim to some of that precious impact.
Then, after the new review cycle starts, priorities change and you never have to follow up on the holes in your data. Or, you can even admit it has holes, you already got credit.
> The employees are incentivized to do so and this is what they measure, so this is what they build
I also work in data teams and this is directly my observation too.
Closely related are Conway's Law (you ship your org chart) and Goodhart's Law (commonly paraphrased as 'When a measure becomes a target, it ceases to be a good measure.')
In Netflix's case I certainly feel the shipped product is too close of a reflection of org structure and incentives and user experience suffers as a result.
This feels particularly evident in the treatment of thumbnails and title descriptions. Often I find myself clicking through on a thumbnail in order to read the description only to find that the I've already read the description and didn't want to watch that title, but the thumbnail has since changed.
I'm sure there is some team in Netflix whose sole purpose is increasing thumbnail clickthrough rates. And they are probably succeeding in that respect by changing the thumbnails. They get to win at their portion of some funnel, even if the net result is a lousy user experience.
To my mind, it's the title descriptions that are the most clickbaity. I get the sense that Netflix are beding over backwards to ensure that you can't form an impression of the flic from that description.
I cancelled my Netflix subscription. Less and less of the titles are well-known 3rd-party productions; more and more are Netflix productions that I have found to be rather dull. Their ability to "divine my preferences" was never much good, but it's got worse.
Of course, they don't really care about my preferences; what matters for Netflix is what they would prefer me to watch, apparently.
Youtube's ability to discern my preferences also seems to have nosedived over the last year. I watched one programme about a certain WWII aircraft; YT then started pushing at me an unending diet of aircraft restoration docs and amateur historian channels.
>> But behind the scenes some of us had this thought that our observations only amounted to short-term gains.
I'm convinced this must be the case at Facebook. I used to love the newsfeed but it has steadily declined so much for me that I stopped using it. It used to highlight a wide variety of friends updates. Then, only starred "close friends" and now mostly junk news items. Amazing things happen for friends (weddings, births) and I dont hear about it (no they aren't blocking me, i've confirmed).
I can to individually to each person's profile and see everything, but how realistic is that.
I'm sure there was a short boost as people checked more and more to news that didn't appear. Now, you check five times, you get the same 30 newsfeed items over and over and then "you're all caught up".
I can totally see the short term boost and long term decline with such a strategy.
Online Analytics is now really nuanced - you need to know what metrics are important for * your * business, not just use boilerplate kpis.
For instance, an e-commerce website is clearly looking to lead the user to a lot purchase and the more they purchase, it’s good for the business so kpis around sales conversion help and recommenders help your business increase sales.
For Netflix though, the users have already paid for the service after which they land on the website. Most users I imagine then expect to be provided all that Netflix has to offer in an easy way. So if I was a Netflix product owner, I’d be more interested in Kpis around search-ability, having an anti algorithm that “suggests” completely random obscure shows, “switchability” of users - how less of a time do users spend on a movie or show.
I imagine they’re doing this but as a user I don’t see this at least - they show the same old stale recommendations for me, I’m always trying to hack their search to find what I want and they continue to invest in content that’s mostly miss than hit. I wish they at least had a directory for me to browse through (at least I’ll be driving their engagement metrics to help them drive their valuations)
Right - but why would they assume I want to watch the "top" show? I've already paid for it and they benefit by having me browse and allowing myself to find that I like right? And what I like is not necessarily what they think others might me like - what one watches is highly variable (can change day to day, even time of the day) and influenced by so many factors that are personal.
Netflix is now a major player in the entertainment business, and I would guess there is a pay-to-play element behind the scenes as well as ROI on new productions, forcing them to jam these options down our throats.
User viewing preferences seem to be a secondary consideration.
I wonder about that too. Of course, the Netflix in-house productions will be getting extra promotion; but setting those aside, I fail to understand why Netflix promotes (e.g.) romcoms to me; I don't watch romcoms. I've never given Netflix any signal that I favour romcoms.
I can only suppose that some motivation other than giving subscribers what they want must be at the root of it.
> i.e. a newsfeed algorithm that had better engagement.
Sometimes (but rarely) I click on something because I ask myself "Why would these a---- show me this? I want to know more about what's behind it in order to understand what could have made their algorithm chose me as a candidate for this content". Or to check where the scam-URL would lead me to, because this also happens, that ads sold on Google and Facebook are scams, links leading to malware.
Congratulations to them, if they think that this is the kind of engagement which should be valued, that any kind of engagement is a good and healthy engagement. These companies are so rotten and their engineers just don't care.
There were tools to help answer your question (predictably called “Why am I seeing this?” after the button copy). Those had so little engagement they were not worth maintaining.
I’ve worked on similar experiments (starting about Stories) and we included long-term holdout experiment to measure the compounded impact of sharing. The fear at the time was more about professional content (with millions of views) vs. your friends. The worry was because people engaged more with the former but posted more if they saw the later.
I left before conclusions were drawn, but (according to press reports) those experiments changed the goals to more leading indicator of long-term trend, like posting rather than Likes. I joined again more recently and expected to see that it influenced the company.
I tried out TikTok for a few nights. It very, very quickly got repetitive. I think the short-form content sphere can only fit so much of an arc in what little time it has. And getting people to sit through it requires being a bit click-baity. And everything was obviously staged, and a lot of times it seemed like...
Okay, so when I was a kid, we made stupid videos too. But we never expected anyone to watch them. And they were often 5-10 minutes long in the end.
On TikTok you can tell that the priority of being filmed is to be seen. They aren't just having fun. It's like TV, and weirds me out.
Off topic: To me they seem to focus on quantity over quality. Can you recommend some of that quality content? I always quit after browsing their catalogue for 20 minutes or so.
Depends on the change and effect size, also on product. Typically somewhere between 1 week and 1 quarter. Almost always there is a backtest after shipping in which a small percent of users do not receive the change for 1-6 months (a "holdout")
I recently got a very gracious trial period for Youtube Music Premium, so I'm giving it a shot. It's UI is what Spotify used to be, before they crippled it nearly completely for my use cases. I'm seriously considering switching because of this. I'm just hoping that YT Music can replicate Spotify's one killer feature - Spotify Connect - before my trial is over.
Definitely has. I remember when podcasts first showed up. Spotify isn't my podcast player so that is pretty much useless to me. I looked through the options for how to hide podcasts and couldn't find anything. I wish I could at least move the podcasts section to the bottom of the page.
The algorithm seems to have picked up the fact that I like watching SciFi and Zombie movies, so my Netflix homepage was this dark place filled with Sci Fi and Zombie movies.
One problem with this is that Netflix only has a handful of good SciFi and Zombie movies, and I've seen them all. So the homepage was filled with movies I've seen already, or the dime-a-dozen copycat movies that just rehash some ideas from popular movies in a slightly different way.
The much bigger problem is that even though I like watching Zombie movies, I actually enjoy lots of different movies. But somehow the Netflix algorithm only ever shows me this one genre.
So I cancelled my Netflix subscription, and went back to occasionally renting a film on iTunes or Amazon. I watch less now, but I end up watching more diverse and more interesting films.
> So I cancelled my Netflix subscription, and went back to occasionally renting a film on iTunes or Amazon.
I did that too, and then went one step further, because some fifteen years down the line, online video rental still hilariously, bafflingly sucks.
Twelve, thirteen years ago I was still renting physical DVDs. Back then, renting a physical item that was produced halfway around the world, shipped to Germany and distributed by a company with hundreds of physical locations staffed by employees was around two and a half times cheaper than downloading a file via iTunes. Apart from inflation, those prices never came down.
Adding insult to injury, the more expensive download is almost always worse:
On the majority of DVDs I get both the original and the dubbed voice track, plus subtitles in English and German.
That's important to me, because my partner vastly prefers either the dubbed version, or at least the original with German subtitles. When I watch something for myself, I vastly prefer the original - sometimes with English subtitles.
In online video rental (or purchase), I can often only get the dubbed version. If the original is available at all, it is sometimes another item to be bought separately. Either case almost never features both German and English subtitles.
And as the icing on the cake, to this day not all product pages on iTunes even _list_ the featured languages of a download, let alone their subtitle languages.
Combine that with all the other indignities of buffet streaming, such as titles being constantly rotated out or the incessant advertising on Amazon before every episode.
So — I've gone back to buying used DVDs. And since they have their own problems such as unskippables and horrible menus, I'm currently looking into building a NAS and will be ripping them into a personal media collection sometime in the future.
And then I will have come back full circle to 2005.
You can also do exactly this with Blu-rays and get really stunning above-streaming quality, though of course the selection and pricing isn't as desirable as DVDs.
What I'm saying is invest in a USB or internal Blu-ray drive, the extra cost is worth it.
I've noticed that streaming video quality is really bad, actually. Especially when the scene is very dark and grey, like Dark, you can hardly see anything because of the compression. Physical media thankfully does better.
You're right, I probably should.. I'm just a little bit anxious of the extra fiddling required to find decent compression settings for BDs, since I've got no experience there.
Thanks for the MakeMKV suggestion, I'll look into it.
VLC isn't an option at the moment, because we've hooked up a 60€ player to her television.
Something with a remote with physical buttons that boots up in two seconds flat, never updates, always works the same and needs zero network connections.
You have described a Raspi media center with a remote and the wifi turned off. It may take a couple seconds longer to boot, but you get back VLC et al in exchange, and there are many good, free options to use it as an interface for your NAS.
I've found that when I watch a Youtube video, at least half or more of the recommended videos on the right are videos that I've already seen before, often music. Does anyone know if there is way to turn this off so it can recommend me strictly new videos? It's infuriating.
The best guess for why this is that I've seen online is that it's because they're targeting younger age brackets more aggressively and kids love to rewatch the same things over and over.
+1 on renting films with some level of purpose. A lot of people I know refuse to pay for movie rentals.
I've seen this too many times: endless scrolling to try to find that one decent movie that comes free with the subscription.
But movie studios aren't stupid. They don't just give away their best movies for free. So those who simply aren't willing to pay are left watching Hitman or The Quake (fine Hulu content).
The price of movie rentals might seem high, but life is far too short to waste it watching things you don't even like that much.
Also: iTunes in particular happens to have very decent staff recommendations (i.e., actual human curation). I wouldn't be surprised if movie rental services like iTunes understand that their customers are looking for quality and not quantity, otherwise those customers would be on the streaming services.
My issue with renting has always been the ridiculous upselling for quality. Advertise 4.99 for a movie, but oh, you don't want the "standard definition" that hasn't been standard since 1996? Well then it's 8.99.
Edit: Just gave Play Movies another chance and it looks like perhaps they've finally stopped doing this.
In the case of Youtube I've basically given up trying to get decent recommendations. At this point I just hope it keeps showing videos of the people I've subscribed to.
I'm in the same boat as you. My question is, does this hyperoptimisation of preferences actually work for anyone other than children? I guess Netflix thinks it does.
When I found myself spending 15+ minutes looking for content and not finding anything, I just cancelled.
I just don't see the value in Netflix anymore. Everyone has wisened up to the fact that creating your own streaming service is extremely important - especially when you consider ROI.
So what differentiates streaming services is the catalog. Netflix now have to compete on the quality of their originals. They have some good shows once in a while but the their lead is not groundbreaking at all anymore.
I have a similar problem with amazon music and spotify: Yes, they are good at picking music I like, but really bad at adding something new.
Spotify is better at this, as it does not give you the one "your radio".
At amazon music it basically played only those songs I told it about, and a few very very narrow matches, but close to nothing I did not already know.
Nothing I've ever seen did that better than Pandora. Unfortunately I had only sporadic access via VPN because of their inability to secure international licensing, but during those times I discovered more music than ever before or since.
Apple Musics "Create station from track" feature comes very close, though. Back when it appeared I remember reading they had humans tag and classify tracks manually, just like Pandora, which may explain the similar results.
I do something similar, I have my main profile where I watch everything and then a "Random" profile that I regularly clear all viewing history from (if I watch something there instead of switching back to my main profile).
The problem is, as annoying as Netflix's algorithm is on my main profile, the algorithm to shovel shit to users it knows nothing about is even worse.
I honestly don't understand why Netflix had to get rid of the ability to literally just scan the full list of titles, optionally filtered by genre. The home screen is an utter failure, I can never find "My List" or "Continue Watching" because it's a heterogeneous mess and they jump around all the time.
Interesting, I wasn't sure at first how it was adjusting the title, but it seems like it lower cases it. Interesting demo, but I think you could do better by just aggressively not clicking on videos like that to begin with. I never see that style of click bait anymore, as long as I'm logged in.
Luckily they can't send different thumbnails to different viewers. Yet.
Veritasium actually did an interesting episode recently about how the major YouTubers optimize their titles and thumbnails using real time data now to figure out what works to go more viral. And apparently there's a convergence on "shocked" face.
LinusTechTips had mentioned that he did an A/B test with "no YouTube face" and "with YouTube face" finding it had a significant impact on the number of viewers.
I suppose we're hard wired to look for shocked expressions.
I was having a conversation about this with a friend last night. I saw a video on plastic CPUs on his smart TV's recommendations, and searched the topic instead of clicking on it because I didn't want to encourage what I called 'Reddit face'.
Is there some Youtube thumbnail generator or a guide? They all literally look the same. Have the same kind of thick white outline around the person and they seem to use the same fonts too.
Even somewhat respectable channels do this thumbnail 'hack' because they claim they've A/B tested it and the "I am yelling loudly" look perform better. Possibly because more children click the thumbs than they otherwise would.
But, having watched the 'crate challenge' that is hot on IG and TT, I'm thinking my thumbnail comment might be giving adults too much credit.
freetube client allows you to control how thumbnails are displayed; default - youtube one or manual fetch frame from beginning, mid or end of the video
My version of not-logged in (which is the default) is far from that. It's regional - but again, unless you logoff on purpose, the selection is heavily based on your choices of 'creators'
The fact that Netflix can adaptively adjust their clickbait to suit each user is pretty stunning though. That's definitely a step beyond "one thumbnail to rule them all".
> users consider each title for a whopping 1.8 seconds
Users are most likely not “considering” them at all. We’re not machines going over one tile at a time and generating a score. Most of the time you’re looking for something specific and just trying to find it.
This type of metric is the worst. I really hope this data-driven fad dies down and we start designing with human factors in mind again.
I'm not certain what you mean by human factors, so correct me if I've misinterpreted.
If by human factors you mean basic heuristics based on human intuition, that fails too. It might satisfy 50% of viewers but there's too much diversity in taste for this to be a good model.
I'm not defending the metric as described, but there's obviously good data driven approaches and bad ones. The trick is to find a solid proxy for the business problem and to try to weed out any dark side effects by applying it.
"Human factors" could include doing something as radical as observing how someone uses your product (no, not with a computer - just watch) and then - and this is the shocking part - speak to them. Ask them why they did that or what they liked and didn't like.
What fantasy world are you living in where you think Netflix and Facebook don't already do this? They constantly run paid interviews with people to source this information.
The issue is many-fold: 1) formalising multiple half hour interviews into something you can A/B test is non-trivial, 2) given the logistics of these interviews the sample you end up with is small and therefore bias.
I worked for the company running the most A/B tests in the industry for most of my career.
Having a user testing session is relatively rare, saved for special one-off products/features. Actually having feedback from those sessions distilled into product feature proposals, and getting them into the product roadmap is even rarer.
In the meantime, hundreds of A/B tests based on metrics like these would have been ran and resulted in a permanent change to the product. It’s just so much easier for everyone involved to trust “the data” vs something that requires nuanced interpretation.
Would Netflix have to do this if it actually had content people wanted to watch? It's just sad watching it double down on this pulp because every other copyright holder took their ball and went home.
A big issue for me is that Netflix now has so much content that everyone watching different things and there's no longer "the show" that you have to watch to talk about with your friends. Instead, there are a handful of good shows every year spread across all platforms and it's come to the point where if I'm just going to watch things on my own, I might as well just watch YouTube.
A few years ago, I subscribed to Netflix. That was it (no cable, no other streaming). The amount of good content was reasonable.
Now, I have to cycle through Netflix, Hulu, HBO, Apple, and Amazon. I unsubscribed to most of them most of the time, and resubscribe as a show or movie comes to my attention (Dune will get me resubscribed to HBO for a month or two, etc).
I still spend less on streaming than I did on cable, but if the whole ecosystem gets any more fractured or any more expensive or any more annoying to navigate, I'll probably start cancelling and just find something else to do.
Unfortunately, finding books isn't much better, or I'd likely spend more time reading than watching TV. But a bad book, for me, is 100x worse than a bad TV show.
I continue to subscribe to Netflix not because I'm thrilled with their selection but because there are a few things I want and it's the business model I want to support. Unlimited watches, no ads, I pay for access to a library. Fuck everything like Hulu and Prime with paid subscriptions that still include ads.
Maybe you should treat books like TV then: give it a half hour, and if you're not hooked, close it and find another.
I know many people say books take longer to enjoy, but a bad TV series can take ~12 hours per season. Most people can probably finish a novel in 12 hours.
Maybe I'm going about this the wrong way but I feel like books are more comparable to movies. A bad movie lasts under 2 hours and I can usually skip around without losing much context because pictures and video convey a lot more information than text. I can infer things that I miss based on background elements whereas in a book if the characters go to a different ___location, it's hard to know that unless I read the part where they go to said ___location.
That's been pretty much my conclusion as well. I can mostly find some kind of long-form content to consume and although the feed algorithm sucks, the recommendations on the video page itself usually make up for it.
Netflix REALLY went for quantity over quality with their library. I'm certain that there are some great shows in there, but I have to start ten different series just to find one I like.
On the other end of the spectrum, the Disney+ library is rather small, but I find at least half the shows to be enjoyable, so that's where I turn to first.
Hm, I really don't understand the approach of choosing shows/films based on what the algorithm tells me I'll like. Watching a series is such an incredible time sink that it's necessary to gather more information in advance of choosing.
I've found a much better model is friends or critics' recommendations cross-checked against Rotten Tomatoes and IMDb ratings. (Which implies you need some people to use the scattergun approach, but I'd rather not be the guinea pig).
This way it doesn't matter if good Netflix shows are "needle in a haystack".
Netflix started out with a huge amount of content. I'm sure they're scared that no longer seeming to have everything will hurt their subscriber base. One way to make it look like you still have everything is to have an overwhelming amount of stuff.
I don't really feel like there's much quantity, either, though. The stuff they're doing is just so very stylistically homogenous.
I guess it's down to how you frame it? Some people see 100 different flavors of ice cream. I see 100 different flavors of chocolate ice cream. Both perspectives are accurate. Which one is more useful to you perhaps depends on how much you like chocolate ice cream.
I also stopped Netflix at some point early in the pandemic. More and more I was seeing good shows go away and nothing decent replacing it. Then they had all these weird shows (Cuties?) coming around that left a bad impression.
Ultimately my canceling of Netflix went along with canceling big tech as my disillusion has grown across the board:
* Amazon (this was hard to cancel, my prime subscription is still paid for the next 6 months)
* Facebook (still use the messenger app to chat w/ friends though)
* Google (search and Chrome - but still use gmail of course and at work)
I'm going to cancel something from Apple to do a full de-fanging of FAANG :)
Actually here's what I'll cancel: Apple Arcade. Everything they release now is just a "+" game from a non-Apple-Arcade title. "Somegame+" where it means no ads and maybe a little extra content. Originals on Apple Arcade are now harder to find and everything in my new/recommended/trending/etc is a "plus title".
Come on, this is beyond hyperbole and absolutely absurd.
Yes, Netflix has content people want to watch.
That’s not a fact up for debate, it’s is absolute certainty and proven beyond all reasonable doubt. To suggest otherwise is on par with arguing that the moon doesn’t exist.
And this is the obvious answer that they somehow manage to keep overlooking.
Just cancelled my sub and won't be activating it again until Christmas. I should then get a decent month of viewing before I cancel it again.
In the meantime I'll be paying amazon much more money for stuff outside of the Prime offering that I want to watch and am willing to pay for (albeit grudgingly).
Nobody was ever going to make all TV and movie content available for $9 a month to everyone. The idea that Netflix could last the way it used to be was ridiculous.
Yeah, netflix saw this balkanization of streaming services coming and they started spending other people's money on original content so they couldn't have the ball taken away from them. It seems to be working for now - they've had 619 emmy nominations - but when there's 15 services that each have 1 show you want to watch, piracy will flourish
> It seems to be working for now - they've had 619 emmy nominations
I canceled my Netflix sub. Maybe this is why? Netflix became desperate to be legitimized by "big Hollywood"/"big entertainment". It used to be the new thing that didn't need to conform with its titles.
Copyright infringement makes everything ever created available for $0 a month. If streaming services can't do the same, they suck and are failing to compete.
When there's only one service to pay for or even two or three most people don't have a problem with paying a fee to have access to everything they want to watch.
But when there's fifteen services to pay for, even if they collectively cost less than the above example, many simply give up and turn back to piracy.
Conveninece also means different things for different people. For me is includes being able to re-watch my favorite content whenever I want and on whatever device I want - no exeptions - even if the the internet is down, even if the creator no longer wants to make it available. The only thing that can guarantee that is something that gives me DRM-free files that I can manage however I want. Unfortunately for movies and shows noone provides this - best non-piracy option are physical discs but even those are so filled with anti-piracy measures (and other questionable technical decisions) which only serve to make the experience worse than piracy.
Theft is not free, copying data is. They're the ones making everything worse for everybody because they can't deal with that reality. These streaming services can't even manage to produce a video player with decent controls, we had better software in the 90s.
Also, copyright infringement is not theft. It's copyright infringement and it's only a problem if you accept the notion that copyright is a legitimate system to begin with.
Yeah. They reap all these benefits. When it's OUR turn to exercise our public ___domain rights, they change the rules of the game. They move the goalposts.
Why respect their rights when they don't respect ours? There is absolutely no reason to recognize their "property" as legitimate.
When theft has a better user experience that’s important. Important because it’s more user oriented than the plethora of poorly designed apps with dark patterns to piss me off but keep me subscribed.
It’s important because it’s the reality.
Today, my janky RSS feed setup dumping stuff to a network folder is better for discovering new shows and movies. I’m more likely to see new content released through my tracker, browsing in 10 seconds than spending minutes scrolling through Netflix, and Hulu, and Prime. Just to see if there’s something new.
These services make it hard to use. I’d pay extra if Netflix just gave me an RSS feed and let me filter it. The fact that pirates are able to do this, volunteering is only part of it.
I don't know if that will be unrealistic in the future. Once AI is making the content you consume, the price seems like it would rapidly approach zero.
It's the only realistic position. Heck, it's the only reason streaming services can exist; without piracy pressure everything would be pay per view or pay per stream at far higher rates.
i bet eventually they will be doing cross platform streaming rights deals, only one i can think off that might never do that is Disney. Your content can only be consumed so much on a closed platform before it losses its profitability and then you just sell its rights to bring extra profits.
otherwise i can't see how this will be sustainable this is going to be more expensive then tv channels were . Yes the content have much better production value but who will be able to afford so many stream services other then just people hopping after consuming what they wanted to watch.
maybe there will be a pay to watch service that has everything and you just pay for the right to watch the specific content.
"Video streaming giant Netflix had a total net income of over 2.76 billion U.S. dollars in 2020, whilst the company's annual revenue reached 25 billion U.S. dollars" [1]
Yes, they can last the way it used to be. Well, they did slightly increase the price because they were forced to spend ridiculous amounts of money on their own content, but they are doing just fine.
Storage and delivery wise its much chaper than YouTube (which is 100000 times bigger in storage needs due to user generated content), and makes all this available for $9 without ads...
The way Netflix presents thumbnail images custom tailored to the user is ingenious. There is a lot of insight there on how people choose content based on preferences for star actors, subject matter, tone etc.
Personally however, as an avid film viewer, this massive algorithmic curation is completely not for me.
A cool example and complete opposite of the Netflix approach is the Mubi approach. There, the focus is not on giving me exactly what I think I want, but instead on offering this narrow curated selection along with content written by actual film critics. As a result I watch things that I did not expect.
This curation aspect is something Netflix strategically completely opted out of. And this makes sense - their goal is to have an active subscriber base and achieving that goal doesn't factor in the existing film/tv culture.
Personally although I do see “sexy” titles on Netflix strewn about, it isn’t a big distraction or problem the way this article makes it out to be. The real problem is that most of the content is B tier straight to video content. I’m drowning in choices and I don’t mean within Netflix but elsewhere - I’m simply not compelled to spend my valuable minutes there and these days I rarely load their app. I’ll probably end up cancelling it, as my household has increasingly returned to analog entertainment (books, conversations) over the pandemic, with other viewing time going to content we deliberately seek out rather than content we casually surf. At one point I was hopeful Netflix could produce first party content that was great. But each promising attempt (for example House of Cards, Marco Polo, or Altered Carbon) fizzles and gets cancelled after a season or two. Today I can’t see myself investing my time into their new content because I don’t want to be disappointed by their eventual cancellation, and I am left wondering who their service is for.
I feel the same. I think their content is a sea of mediocrity.
I cancelled my sub and I make do with Amazon Prime which is not much better but does have some good exclusives and obviously the shipping benefits.
I’m also considering a return to piracy with a NAS since a lot of movies (esp. classics) are simply unavailable anywhere, even pay-per-view services like Amazon, Apple or Google.
I am wondering if this is what getting old is like. I get more enjoyment out of rewatching all the stuff from my youth than new content. And I am only mid 30s, but I would rather close my eyes and pick a random movie or tv show on my NAS from 80s/90s/00s and watch that over new content.
I really hate it that when I'm digesting an episode of Rick and Morty I have to scramble for the remote or I'm watching some completely uninteresting series about a family in the 19th century or something. I just want the end screen and sounds to finish and then turn off the telly and go to bed. In stead I feel slapped in the face.
Consistently my watching experience ends with (increasingly large) negative feelings. How can that be a good choice for a company offering watching experiences?
You can now turn off autoplay of both trailers and next episodes, in settings. (Account menu -> Account -> Profile and Parental Controls -> [your profile] -> Playback settings -> Change)
Take a moment to think about having to go through six layers of menu tree to access a frequently requested feature. Now imagine explaining this step to someone who doesn't intuitively understand computers and whose children bought them a Roku for Christmas.
I'm not fussing at you, just thinking about this first-world hell.
I believe Netflix has talked about thumbnail experimentation at length in some of their engineering blogs [1]. To me, it seems perhaps Netflix has figured out a running theme in the author's viewing habits... perhaps the scandalous, soapy stuff is what they keep clicking.
Same kind of reaction I get when a guy friend tells me they keep getting ads for women's bikinis or women's underwear in their Instagram ads. It isn't sexualization of Instagram... it is a reflection of your interaction with the platform!
That response to complaining about Instagram is becoming a frustrating meme because it isn’t true that Instagram doesn’t offer sex appeal to the uninitiated, and rather than discussing that issue, this simple response just shuts down criticism of the social network by embarrassing the person who brings the issue up. Lest you turn the meme on me, I’ll say don’t have an account.
I don’t think this response is dismissive: it’s a technical explanation of what is happening. Understanding it helps us find a solution.
Netflix (of Facebook) rates content along embeddings, and highlights content along embeddings that users engage with. If a given user engages with clips featuring people in bikini, that system won’t pass the judgement (one that the author is passing, a judgement you dearly regret, ‘Lest…’) — but one that the commenter didn’t pass. No one in that process (at Facebook, Instagram or Netflix Tech team) is choosing to make things sluttier, or asks someone to bend over a little more (Netflix being a film and series producer too, I’ll reserve there).
Given that context, if, like the author, you believe that the content should not engage along certain dimensions then there’s several possible takes:
* a paternalistic approach to downgrade dimensions that are deemed unhelpful; that would have been popular in my Catholic school, less so on Hacker News, I’d expect;
* a empowering approach that either:
* name the dimensions and let users downgrade them; that’s harder than one would expect for many reasons, but roughly: naming is power and embeddings are not human-legible;
* let users downvote some recommendation, even if they’ve engaged with it.
Both Facebook and Netflix but also YouTube, Google Search engine, all have strangely avoided building those negative feedback for click-baits. It many ways, it’s a mystery to me, especially as someone who built so many recommendation systems. I’ve always used implicit or explicit negative feedback, to just engagement and the recommendations were meaningfully better.
I’d love to hear from people who have explored that approach if there have encountered issues implementing them. I can share two details that are not sufficient but I hope are relevant:
* The biggest problem at Facebook has always been that Likes were performative: your friends saw them, by design; people refused to use them as a way to improve their recommendation because of that. When we discussed a drop in posting, or when the five other reactions where debated, there were several suggestion to make a parallel form of Like invisible. I’ve seen those ideas being shot down repeatedly; I was never told why. I assumed there was a fear those would cannibalise the public Likes, and those drove a lot of engagement opportunities, but that was speculation on my part.
* Most effort to understand recommendations (“Why am I seeing this?”) had an abysmal engagement rate. So much that maintaining it wasn't worth the effort.
Nice response. Unfortunately, the original response is dismissive even though it’s about a technical topic, but it wouldn’t be if discussing just about any other content category and you’re right that those kinds of mechanisms are worthy of analysis.
I haven’t worked at FB but don’t metadata-adding contractors training ML models serve the function of invisible, more complete likes? My understanding is all these large social networks employ or employed human-created content analysis. And then the right amount of noise and divergent suggestions can be added to serve the goals of the network.
The issue rarely is whether the content can be interpreted properly, or at least hadn’t in the last few years. There are contractors flagging edge-cases, issues around nuances like photo-realistic first-person shooter video game footage (common) vs. footages from actual fire-fights (banned since the Christchurch attack) so it _is_ and interesting technical question but that’s not what is challenging to improve the experience.
Some people find others talking about their success inspiring, others find the bargain obnoxious, others still feel demoralised by it. Those can be the same viewer at different time. Without a better way to let people express their reaction than Likes, Comments and dwell time, you won’t be able to tell what’s the long-term impact. “Blocking Facebook” is the only external solution, and getting no-signal isn’t going to help Facebook improve. Every idea under the sun was suggested internally, but whether they are given a chance is a delicate question.
A friend of mine was hired to do design work for Netflix - he ended up having to create tens of different thumbnails for every show, which netflix would then run through their algorithms and decide which one to show to which user.
I've looked at my netflix page and it seems that some silicon valley psychologist decided that they should be showing me thumbnails of guys with sad faces looking at something far away. Almost every single thumbnail looks like that.
Sometimes I wonder if the engineers behind the scenes even realize how creepy it is what they are doing.
Does your friend work with User experience researcher? I know it sounds ridiculous to delegate “this looks creepy” to an expert, but that’s where those people are valuable: they can enrich an impression with detailed feedback.
Also, if your friend has any sway, can you beg him to suggest a “Stop recommending this to me” button, if they haven’t. The platform is better at removing movies I’ve stoped watching from the “Continue watching” selection after a week but… that’s still a bit long. I just want the agency of telling the machine No on occasion.
And if that works, thankfully, we can get a “stop showing me the same image” button too.
Netflix has literally zero discoverability if you want to go outside what it suggests. If I click stuff that initially seem appealing and then based on seeing decide I dont like it, tough luck. You clicked on it, because you liked name and pic, turned out different thing that you like, but now netflix is forever convinced you want this.
I tried to go through documentaries category, because I was in mood for documentary. Pure hell and every time I viewed movie detail, it jumped back to start.
I dont have an option to click on something else and Netflix does everything possible in order to prevent me to find stuff I would like. The result is that I dont idle watch netflix, because it is too much work and we watch only shows someone else told us about and we searched by name.
I’ve been a Netflix subscriber since the DVD days, and for the last several years all the best shows I’ve found, I read about elsewhere first.
Which is kind of insane, but I got used to it, and I will google around for interesting new Netflix content instead of bothering with their scroll-fest.
I unsubscribed many years ago but occasionally I am tasked by my significant other to find something on Netflix (shared account with her family).
I actually have pretty good luck with searching for random things. Like I searched for "cool" the other day and found something good that would have never shown up otherwise. Seems like a way to substantially bypass the recommendation engine "smarts".
Yep. You can also search by genres including their ridiculously detailed sub genres or cross-over genres.
From what I’ve observed, though, normals consider TV such a “lean back” they won’t even go to the trouble to pick Movies or TV before browsing thumbnails for 1.8 seconds each…
> It isn't sexualization of Instagram... it is a reflection of your interaction with the platform!
No, that "gotcha" is blatantly wrong. If you follow running people, Instagram will float you popular running content. What's popular running content? Girls posing in yogapants. Same for other niches. The most popular and thus shown content will always be a sexual version of the theme.
Yeah, sure. I like desserts. I’ll eat a tasty looking one if presented with the opportunity. But it doesn’t mean I want my kitchen or living room filled with desert ads.
While thats certainly part of the problem, I distinctly remember that, when I created my instagram account and followed the first few accounts (all miniature painting stuff, I use insta for exactly nothing else) the suggestions in the search where sprinkled with stuff not fitting in that genre. Like car content and barely dressed women posing. I had to do the "long tap" => "not interessted" dance for a few days to get rid of most of it.
Sure, maybe its something the algorithm connects. "This fellow likes painting small figures, other peoples who do that also like cars and women, so he will like them as well" but its harsh to say "you only get what you asked for" in this context imho.
It might be a reflection of my interaction. But I distinctly remember telling Instagram’s Discovery feed to stop showing me sexy women. This would work for a few weeks until the women would eventually come back. I’m not sure if this is an issue of the request (stop showing me X) expiring, or if I lingered too long on some other photo one time, but either way it wasn’t content that I sought out. Instagram kept finding a way to recommend it.
Notably, I kind of liked the Discovery feature when it was showing me relevant and unsexy stuff. But I’ve since deleted Instagram.
> The company seems more brazen in its strategies, more willing to promise something and then absolutely fail to deliver, often using headline tricks familiar from the social web. That’s what clickbait is: luring someone into clicking, and then delivering something other than what the headline made them want. [...] my homepage illustrates The Big Lebowski with a photo of an angry John Goodman pointing a gun, as if the Coen brothers’ comedy is actually some kind of revenge thriller.
Netflix did not figured out how author viewing habits. It is trying to figure out author clicking habits and then disappoints. It misleads author into clicking wrong stuff.
> Following the success of the comically generically titled Money Heist, Netflix recently debuted the new, even more generically titled hit Heist
Because the recommendations are based on superficial but illogical similarity.
Movie posters have used salacious and sensational words and imagery for as long as they've existed. If anything, Netflix titles have been slowly working toward parity to that standard, as opposed to devolving to actual clickbait.
I see your point, but movie posters didn't live in your home and didn't have the power of mass personalization.
An analogy might be junk food. Sure, it's advertised in all sorts of tempting ways in supermarkets, but imagine if your refrigerator suddenly gained the ability to show, prepare, and serve you customized/personalized junk food whenever you opened it.
Movie posters were ubiquitous in homes, a favorite of teenagers, often the racier the better.
I see your point, but I don't see a problem with Netflix content using sensationalism to drive clicks. In fact, it's probably one of the vanishingly few legitimate uses of that type of visual manipulation. They pull visually stimulating stills from the content itself, combined with text that gives you a quick indication of what's in store. I take issue with low effort or misleading titles, but it's a constrained medium in a commercial context. Not a lot of ways to vary the interface.
Yeah this makes me feel so old. This is not really different than scanning the box covers at a Blockbuster store or, as you mentioned, movie posters which have probably been misleading for nearly a century.
Let's be real about Netflix: after looking like it was going to disrupt the media conglomerates, I now think that all it has managed to do is wake the sleeping giants. Long-term, I am bearish on Netflix. I would go as far to say that Netflix is slow-declining its way into being eventually acquired by a member of the big six.
Here are the issues:
First: it's basically the most expensive streaming service, topping out at $18/month.
Hulu's most expensive plan is $12. Discovery+ is $7. Disney+ is $8. HBO Max is $15. Paramount+ is $10. (All prices ad-free plans)
I think an argument could be made that all or nearly all of those services are offering a better content library at a lower price when compared to Netflix.
Discovery+ especially...holy hell if you are into reality shows it's endless. And it's $7. I would pay $7/month just for access to every House Hunters episode imaginable without ads, lo and behold my dream came true.
Netflix is doing this clickbait stuff because their content sucks. Clickbait is what you do when your content doesn't speak for itself.
Sure, every content business has to make a "headline" to draw your attention. But when you see a "clickbait" headline in something like The New York Times you know you're being drawn into something that can be potentially rich in effort, and therefore the term "clickbait" doesn't really apply. At least there's an article behind the hook. "Clickbait" more specifically means you're being tricked into visiting something that everyone knows definitely sucks, including its creator.
Netflix knows their content sucks, and I'm not sure they care or can think of a viable business model to improve it.
Here’s what I don’t understand (either about the premise of the article if incorrect or Netflix as a business if correct): why does Netflix need to drive proactive short-term engagement? They don’t make money from ads. If I’m subscribed (I’m not anymore) then their business is accounted for. I’m subscribed because they’re a better platform than TV and when I want to watch a show they have, I will. They’re getting my money regardless and they aren’t getting more if it the more I watch. Maybe they really feel pressure from other streaming platforms?
The big complaint they get about content is that there isn’t anything that is appealing enough to get customers to start watching. Once people start watching a show they usually are pretty happy. But they spend 30 minutes trying to find a show. This is why they implemented the button to pick a show for you.
In short, they want you to start something. They think you’ll like most of the stuff they recommend to you. The worst case is you click nothing and then go to Hulu and browse there.
Note: Every assertion I made is made up by me. Just my theory.
They don't want you to use other platforms at the same time. Because if you spend too much time outside and start liking them, you will be at risk of unsubscribing. Heaven forbid they need to compete on content quality to retain you. This will be much easier and cheaper.
If I search for "streetcar named desire" I want a finder which says "nope, but here's the other work by {Elia Kazan, Tenesee Williams, Marlon Brando, Carl Malden, Jessica Tandy, Kim Hunter}
Oddly, I don't want "here's the abortive experimental 1970s film shot in Russia by a drug crazed student also called "streetcar"
But.. from Netflix's PoV, its equally likely I did want that because they know I watch e.g. Luis Bunel films
So.. how can they know? Answer: they can't. They simply can't get my mood right, all the time. Sometimes, they will do well and guess. Sometimes, they do really badly.
Another take: How many fans of the british "Office" wanted to be told to watch the American "Office" ?
I do get a lot results that way on Netflix on failed searches… It doesn’t present only lexically similar results (“streetcar”) but thematically related films (classics, of the same era and related directors or cast)
My main issue with Netflix is that where I live their movie catalog (esp. classics), is very short, seemingly in favor of Netflix’s own produced “shovelware”.
Also, regarding a search for “the office”, how would Netflix know not to recommend the US version if you’re looking for the UK one and vice versa? That’s such a specific nitpick…
There are a million nitpicks in the city, this is just one of them.
Netflix and Amazon both have a strong interior belief this smart referral thing is good and who knows: A/B testing may even validate it. I just know that recommendations in books film and music creep me out, maybe it's an analogue of "uncanny valley" but being told "you may also like" kills the mood for me. Dammit, I want to like what I like, not what statistics thinks I may like.
"The office" thing, yes, that's the whole point: ask me Tuesday and I might say yes. Ask me Monday and Sunday, I might cancel my subscription. I talk about this with a lot of people and book and film recommendations are toxic to friendship. You love STNG? Oh you would love Babylon 5 no, just no.
And I don't think "I'm zany" or anything cute. I think this is a normal reaction and a common reaction. I bet Netflix has never asked "do we creep you out a bit snooping on what you watch and recommending things" that literally.
I thought you were talking about search results and not algorithmic recommendations.
I think most recommendations for video services is done through collaborative filtering and similar ML techniques, meaning you get recommended what other profiles similar to you have watched. It may be possible to infer a profile of "people who like the Office (UK) but will not like the Office (US)" but I imagine it may present challenges, especially if the intersection of people who like both is large.
> I thought you were talking about search results and not algorithmic recommendations.
I don't personally see a distinction in irritation or creepiness here. They can't know I don't want experimental art streetcar or a modern remake. So I have to be a bit forgiving.
Amazon do something more loathesome: they proffer "in the style of" before hits for the actual author.
Anyone have a good website that lets you find netflix shows? I like a wide variety of stuff and I think Spotify/Netflix/Youtube always gets too confused to be useful.
Netflix has also become unbearably "woke". I've simply given up on Netflix for that reason but also due to many other reasons like the what is described here.
Not the commenter you're responding to, but I've found quite the opposite. Content on Netflix is increasingly trending towards one narrow "woke" way of seeing the world, and nothing that falls outside of the narrow, American left perspective, or encourages any diversity of thought is available anymore.
Can you proviede some examples of this narrow way of seeing the world? TV shows tell stories. Occasionally those stories aren't going to be towards one's liking.
For example, Amazon Prime is constantly spamming me with ads for Jack Ryan. My take is that they do so because they paid a lot for the IP. Not because it's trying to shove the benefits of American militarism abroad down my throat.
They're shoving the agenda down our throats. A lot of shows become overly politicized, with messages of trans rights, BLM, and other stuff being injected to such a degree that it becomes too much and frankly artificial.
I'm not against political and societal messages in TV shows and movies - that has been done for decades, but not if it takes away from the writing and quality of the show. We see that in extreme levels at the moment.
Is "the agenda" not a bit of a conspiratorial way of looking at it though?
It's been notoriously difficult for creators to find mainstream outlets to tell stories about, to use your examples, trans and black lives, that aren't watered down out of fear of alienating a white, heterosexual audience. It's politicised in so much as defending your right to exist in the public sphere is always political, but is that a problem?
I love that we're starting to see stories that show more perspectives. It doesn't mean every show or movie will be amazing, but when has that ever been true? I don't think it follows that doing this "takes away" from the quality of the writing. A badly written show is a badly written show.
> I'm not against political and societal messages in TV shows and movies - that has been done for decades, but not if it takes away from the writing and quality of the show.
I think if you replace the last part of this with "as long as it doesn't bother me" then your statement is probably more accurate. From my experience bad shows are bad because of terrible writing overall, not because one of the characters is a minority.
> We see that in extreme levels at the moment.
Compared to a decade or more ago when you weren't seeing trans people on TV, I guess you could say it's "extreme" nowadays.
I'm going to front-load this by stating that I 100% agree that representation matters, and absolutely welcome a diversity of perspectives in media. Furthermore, I do believe that much criticism of "woke" media is a knee-jerk reaction to new, unfamiliar values and shifting power.
That said: Netflix's deployment of "wokeness" can feel completely shoehorned in and utterly arbitrary, a cynical technique that affirms the viewers' perspective, thereby increasing engagement.
I'm not "outraged" by it in the slightest. It tends to align with my own worldview. But at the end of the day it just makes for crummy content.
After reading the article on my phone and hitting the back button to get back here, I'm instead presented with a screen showing more content and pleading me to stay on the site (Keep on reading!). Talk about the kettle calling the pot black.
I was very confused by what was happening when I encountered that screen. Is there a way to block the overriding of back buttons, particularly on smartphones?
It's probably there as a consequence of single page web apps needing the ability to store your ___location on the webpage.
Maybe this could have been done in a way less prone to abuse, but basically you can't really trust your browser history to reflect what you think it is.
It's interesting, because one of the earliest lessons I was taught in college was that greedy algorithms tend to lead to sub-optimal results.
Is the problem that they're not teaching that in college anymore, or is it that the people in charge of these things are non-engineers who don't necessarily even know what a greedy algorithm is, let alone that the principle might apply to business practices and not just software implementations.
> "The network’s own research shows that users consider each title for a whopping 1.8 seconds, and that if users don’t find anything in a minute and a half, they’re gone."
It's Netflix's fault this is true!
There's no way to review the content that is actually on the service. I look at an item for 1.8 seconds because I keep being shown the same five promoted shows and I've already decided about them two weeks ago, or already seen them, or whatever. I click off after a minute and half because I'm seeing the same shows and there's no way to find different ones or scroll through all the content.
What these streaming services actually need is a thorough catalog listing with a text description of each show. People will occasionally spend 20 minutes scrolling through that, tagging what they want to see (like the old TV guide). And then on a day to day basis, they'll spend 20 seconds deciding which of the already-tagged-by-yourself content they want to watch today.
All the streaming services are way over-optimized and it hurts them badly. One complete scrolling catalog with lots of ways to search it, lots of ways to page through it, and easy "I want to watch it" tagging.
Maybe a bit off topic but wanted to share: For anyone who wants to watch classic and top films from across the world, please check out The Criterion Channel[1]. Their collection has movies such as Solaris and other films from Andrei Tarkovsky etc.
I dumped Netflix the second it became clear that they were pulling a bait-and-switch, replacing AAA content from major networks with their own material.
Algorithm-driven content feeds simply do not work for me. I don't want to watch TV all day, nor do I have a genre that I watch to the exclusion of all others.
It kind of works for me when I'm tired at the end of the day and I don't want to make choices. So yeah, I'll watch that Caitlyn Jenner documentary or whatever, fine.
It feels harder to find stuff on Netflix when I'm actively looking for something decent, though.
One day I realized that getting dragged into the bingebait series constantly thrown at me just made me depressed. After canceling the streaming services I feel I'm not missing any of it. Also I read more books now. Whatever makes you happy I guess..
[REQUEST] a userscript that replaces the thumbnails with the description of the video/film/TVShow for YouTube and other streaming services.
I feel like replacing all thumbnails with useful text that describes what the video's going to be about would help reduce analysis paralysis for at least some people.
I still use my RSS Reader(QuiteRSS) to "subscribe" to YouTube channels, and I'm never unsure about which video to watch. I think that's due to the fact that RSS is mostly text, and I'm not overwhelmed with images from everywhere all at once.
A good description sells me on a video more than thumbnails ever could.
So my Solution: Replace images with descriptive text on every video streaming platform.
FWIW, this isn't new. Netflix has been showing custom thumbnails for almost a decade now. If the author is getting nothing but scantily clad people, it's because they mostly click on thumbnails with scantily clad people.
I found this article to be so spot on. There’s a mindset often found in tech companies that essentially comes down to “in every decision the data must win.” I see this in my day to day working at a large saas company all the time, and pushing back against this mindset is akin to heresy on some teams.
What the author points out is happening to Netflix is the inevitable late stage of this mindset. To me it looks like the home screen in Netflix is optimized for superficial engagement, at the expense of the actual value and joy customers get from the service.
I think I know why Netflix keeps showing movies that we have already seen. There is just not enough high quality movies that we haven't already seen (speaking for avid movie watchers). So if they skip all the good ones that we have probably already seen, it'll be a sea of mediocre ones with an occasional unwatched gem. Then we will say Netflix has mostly B-grade junk. And they can't have that truth out in plain sight.
Veritasium talks about thumbnail and title experimentation on YouTube, and argues for the need to do so on platforms that surface up and recommend content to users:
Christ that guy comes across as an arrogant prick. Here, "Veritasium", let me rephrase your shitty, overly long, condescending video: "I deceive my viewers because it makes me more money".
I'm not going to rewatch it, because, as you said, it's long, but what he's saying is that there's a difference between baiting and never delivering and bating and delivering.
Also, educating isn't just about putting forth the facts, but also persuasion and spreading. It does coincide with him making more money, but there's nothing bad about that.
Dan just outed himself as sensible to this kind of clickbait: indeed, the Netflix homepage is different for everyone. Try to have a look at your friends or partner's homepage, it's pretty interesting.
It’s also due to a lack of content from the lockdown, however this trend will continue, people will learn to navigate it. For me, I’m going with the dude.
Hollywood, of which netflix used to stand alone but has now been completely assimilated, has ceased making anything of quality in favor of only making things based on the identity politics of the people involved. There hasn't been a good netflix original since orange is the new black and there probably will never be, at least for years to come.
I am not sure I understand what clickbait actually is. It seems to be a term a lot of people use, but in many different ways, and often in a 'I know it when I see it' sort of way.
The author claims that basically any sexiness constitutes clickbait, that a hammer is clickbait, and wield is a gun is clickbait. I think only the last of those is plausible. Title images really give very little information, and usually suggest only what genre the film is.
If Netflix is nefarious, why even put genre labels on? Afterall, wouldn't that discourage people from watching films with outside of their genre?
And the best example they can come up with is that a show about norse mythology has as it's cover art... a hammer? As in the object that we associate with Norse mythology? What exactly did you expect? Okay, no hammers--better show a man instead. 'oh why are you putting a man there when it's about mythology?' You can't win.
I get the impression the author just wanted to beat on Netflix and tech in general.
Take the Big Lebowski thumbnail on his screenshot. While that scene is in the movie, it’s not at all central to the plot. It’s not that remarkable or even show the main character. There is no shooting. But a man aggressively pointing a gun gets you curious, more than the actual movie poster.
It’s clickbait because it is simply picking the image most likely to elicit a reaction - and misrepresenting the nature of the movie in the process. I think the article did a decent job of explaining the psychology behind it.
Another simple example: imagine the title of this post was “Netflix is dying - here are three reasons why”. No actual connection to the content, just something to grab your attention. That’s clickbait.
Yea so the Lebowski one was the only one I really agree with, and even that's is a stretch. We have to believe that the title is actually mistepresenting the genre... Maybe but it's still liated as a comedy.
The other 2 examples did not make any sense to me at all.
I guess to summarise what I mean: of course people want to make good covers. It seems these accusations depend on us really assuming a very nefarious intent. It just feels like any art couod be clickbait by that. Eg. If they used an image if a bowling ball, I could say 'Well, this movie isnt about bowling!'
The main point of the article in not about making good covers. It's about using covers that match - sometimes marginally - your interest even if it's not a main part of the movie, misrepresenting the content of the movie to make you click on it. Clickbait.
The Lebowski example was just that: he likes action and Netflix presents to him the "action" thumb for the movie - even if the movie is not exactly John Wick.
This sounds great at first, and certainly is straightforward if you want a promotion. But behind the scenes some of us had this thought that our observations only amounted to short-term gains. Although we had small long-term experiment holdout groups, the truth is they were rarely reviewed because it was unsexy.
My current thinking is that features like the echo chamber effects from Facebook's algorithms, Snapchat's snap streaks, and clickbait like this, all serve to optimize short-term engagement. Yeah, I want to watch that sexy new show or keep my streak going or have my opinions validated. But there's a diminishing return on clickbait, hollow articles isn't there? I can only fill up so much time with garbage like that before I'm bored. I can only like so many posts before I feel like they're all the same. And once my snap streak is broken I hate snapping.
The data/engineering/product loops at tech companies favor boosting short-term metrics; The employees are incentivized to do so and this is what they measure, so this is what they build. That's why we end up with features like this. That's why Snapchat fell off. That's why Facebook fell off. And that's why Netflix feels increasingly stale (despite there being a lot of quality content if you dig).