Facebook will lose 80% of users by 2017, say Princeton researchers (2014) | Hacker News Reader

Facebook will lose 80% of users by 2017, say Princeton researchers (2014)

theguardian.com

263 pointsNicoJuicy206 comments

> John Cannarella and Joshua Spechler, from the US university's mechanical and aerospace engineering department, have based their prediction on the number of times Facebook is typed into Google as a search term. The charts produced by the Google Trends service show Facebook searches peaked in December 2012 and have since begun to trail off.

Well, yeah... The more people are already familiar with facebook, the less they will be googling/researching it. If anything, considering the popularity of Facebook at the time, it should have been seen as a sign of vitality.

I'm pretty disappointed that the Princeton researchers were so blinded by googlinums that they forget those googlinums don't exist in a vacuum. Probably why mechanical and aerospace engineers aren't fit to make predictions of a social nature (among other engineers, too, sadly).

Not to mention people who use facebook only through the phone app are not likely to ever search for facebook on google. I'm often amazed at how almost-completely phones have replaced "desktop computers" for many people I know (Anecdotal, I know).

I agree with this to the extent that I don't even care if my site ranks well on Google. The traffic coming directly from Facebook is so much greater.

I've been in Japan for the last 7 years and it always bewilders me that so few people own desktop/laptop computers here but everybody is on the internet 24/7. (possibly not only anecdotal...)

I've heard explanations for this involving smaller, more space constrained apartments - and thus less interest in filling it up with yet another machine (tm) and less interest in spending time cramped up at home. This also goes hand in hand with e.g. PC Bangs being more popular over there.

I wonder if this might also go some way towards explaining the salary-man phenomenon. Where employees do excessive overtime at the office. Presumably at the office, they have a work provided laptop/desktop and internet connection. Also the office space may be more spacious in total than their apartments with most of the amenities of home.

Thus they may derive a benefit from staying at the office during the evening rather than going home to a smaller space and no computer.

This is precisely the case for me. I neither have proper space nor the interest to have a computer at home. When I come home from work, spending time on the phone handles everything.

PC bangs popular in Japan? I don't know the answer, but I doubt that since Japan doesn't play enough pc games.

Lot of people in India did not even use desktops because of the high costs they directly jumped to mobile.

Same in the Philippines.

It is possible shrinking PCs/all-in-ones will make a difference in this.

A nicely designed all-in-one PC is a lot more palatable and manageable to the average person than the old tower/cables/monitor/speakers/etc.

I assume people, (even tech savvy HN readers) access sites as a search because it's less effort than urls, typo resistant, and almost all browsers installed now have search built in - it's not a sign of a naïve user.

However, i'd assume for Facebook a much larger % of users are now via smartphones - so get their fix directly through the app, with no search engine involvement.

But your method also ignores mobile access - which is huge.

autocomplete

Looking at one poor analysis and concluding engineers aren't fit for social observations is a bit of a stretch. Our job is literally analysis of problems and formulations of solutions. You win some, you lose some.

You seem to be assuming their conclusion was based on only this data point.

You seem to be assuming that the conclusion is based on the data. This could just be an example of labeling.

Googling facebook is how many people access it and other things. That will probably correlate more to a older and less savvy audience, but should still give a decent proxy in terms of popularity of the web property.

What it _doesn't_ take into account is mobile app usage, which might have been what sucked in all those people vs just total loss.

Even on a desktop / laptop, many people probably use the browser's auto-complete or bookmarks for access.

I just have it permanently open now.

Talk to people who run usability studies and ask how many users know that the browser can do these things.

People joke about "my grandfather opens his browser and types 'Yahoo' into Bing, clicks the link and types 'Google' in to Yahoo, clicks the link and types 'Facebook' into Google to get to Facebook", but it's not that wild an exaggeration of how much of the developed world -- which is not computer-literate -- uses the web. That's part of why walled gardens and native-feeling app experiences are so successful.

Adoption of Chrome likey erases that pattern. As more and more people adapt to searching from the address bar, autocomplete is also used more and more.

No people type google.com into the address bar and then enter facebook in search page.

You'd be surprised how often people do in fact do exactly that. I know two personally. I don't have overall figures but it's definitely greater than zero.

Isn't there a service which shows popularity of search terms? That would be one way to get a handle on frequency.

Found it:

https://www.google.com/trends/explore?q=Facebook

I don't know this tool well enough to know if this is for queries where the only term is Facebook. Happy to be corrected!

I put a link to Facebook in grandma's bookmarks toolbar, she generally just clicks that to get started.

Isn't the growth at the other end?

What browser ? I just click on the internet.

Not sure most people even use bookmarks to be honest. Maybe my non-technical friends are atypical, though.

"Navigational searches" are a thing and they are very optimized by Google. But yes, auto-complete likey handles a lot as well.

Yeah, I think essentially Google used to be the only (or at least most popular) means to discoverability of something on the web. App stores have replaced that for many classes of searching.

Of course, Google's secret is that it owns a monopoly app store on the dominant mobile platform too. ;)

But app usage isn't reflected in search trends...

I don't think this is even vaguely true.

App stores have terrible search features, and return thousands of scams before the real thing for anything except a small class of very well-known apps.

If it is true, I wonder what class of user it's true for, and why they got this way?

Because many users have no idea on how to find out what things are available and what's popular, and/or even that there might be aggregators of this info, or the concept of such.

Observe how many people do not even do a search for very simple things. The concept that they might be able to know or discover certain things doesn't seem to actually occur. Of course, this needs to be distinguished from their just being lazy, but even this is problematic, since it's almost always the case that finding it themselves is much faster than asking someone else (who will often essentially just do what the asker didn't...). It's almost a kind of learned helplessness, but with the "learned" part often being very short or implicit.

    > Googling facebook is how
    > many people access it

I suspect mobile usage of Facebook makes the percentage of people Googling for it a rounding error.

Facebook is accessed mostly mobile these days. I remember that they said sth around 70% mobile usage, and that was 1-2 years ago. Many people never use the desktop app (me included).

Facebook pushed mobile first several years ago and were very successful with it.

The web app seems plain ugly now. the mobile UI is so much better

I will be probably faster in every task you can imagine in the web app than in the mobile app. I don't see how mobile UI is better.

Are we talking about the Facebook desktop website vs. the Android/iOS app? If so:

The Facebook iOS app is snappy on my iPhone 7. Compared to that the Facebook website I access with my MacBook Pro Late 2013 does not feel smooth, and the navigation is terribly broken. Pressing the browser back button often simply does nothing even though the URL in the browser location bar changes. It is frustrating. The iOS app just works.

removed the app+messenger from phone couple of months ago, never going back. but I guess I am an outlier

no, i did the same removing both apps when messenger took over my texts making my phone slow.

That was an optional change :P I have both Messenger and Signal installed, and Signal in place of my normal texting app. It's doable.

I did the same and saw an 11% improvement in battery life by removing those two apps. I had relatively similar amounts of usage initially, but eventually reduced usage to more scheduled times 1-2x per day due to lack of notifications.

I still have the apps and account but never use them anymore.

I can report at least a 20% improvement in stress levels.

Facebook is way too overwhelming for me to use. Fortunately the parts that I need to use occasionally (birthdays and messages to acquaintances) can be achieved by spinoff apps (Events and Messenger)

Or anyone in china

Both Google and Facebook are blocked in China, so there's nothing to measure.

Edit: interested to know why the downvote.

Even better, people google "google" to access google.

https://www.google.com/trends/explore?date=all&q=google,face...

Note that the searches for FB on Google have been dropping while the searches for google haven't really.

I wouldn't be particularly surprised to see Facebook do a face-plant soon given the trends, as I believe google-trends is not an unreasonable proxy for how often people want to get to a site. But, perhaps that's just my optimism for humanity peaking through...

Also interesting to note that they say Facebook Google searches peaked in 2012, which was also the year Facebook shifted their focus to mobile phones according to this MediaWiki article/timeline:

https://en.wikipedia.org/wiki/Timeline_of_Facebook#Product_a...

> mechanical and aerospace engineers aren't fit to make predictions of a social nature

this seems like a hasty generalization. it's probably fair to say that a mechanical engineering education is unlikely to include much depth in topics of a "social nature". (of course, there's nothing stopping a mechanical engineer from diving in, be it formal education or self-directed.)

that said, identifying one's assumptions (such as: a decreasing search trend implies decreasing interest or utilization) is an important facet of any sort of research, be it "social" or not. i think describing this as a "science/engineering geeks aren't qualified to talk about social stuff" situation seems to miss the mark.

Haha exactly my thought....author clearly does not use Facebook or internet for that matter

Little did the researchers realize, the reason for the fall-off in Google searches for Facebook is that Google itself peaked in 2014 and as epidemiological models which are clearly applicable to everything show, Google must have lost about 91% of its users over the last three years, which the researchers failed to control for. However, this is small wonder, given that viral epidemiological data (which is clearly applicable to everything) shows that Princeton itself peaked in 2008. The smart researchers must (clearly) be somewhere else, as clearly nothing can last except for eternal truths from other irrelevant disciplines. I wonder what the next hot university is?

The combination of mobile use, plus autocomplete of browser history should explain almost all of this.

It's ok.. they will just publish another paper in 5 years how they were wrong and now mobile is the answer.

However, that Facebook doesn't have a phone is a problem. People have to download and install the app. I would imagine many Android users google Facebook to install the app.

So then, the study may actually be valid.

Facebook pays several Android device vendors to preinstall their mobile app as bloatware / crapware. Users can disable the Facebook app but they can't totally uninstall it.

Society = Google + O(ε)

An aerospace engineer might be able to rattle off all of the important higher order and relativistic corrections to the hydrogen atom, but when it comes to society...

I think people started realizing that typing fa instead of Facebook was good enough ( I did)

> fa

You get as far as the 'a'?

Facebook actually did a hilarious debunking: https://www.facebook.com/notes/mike-develin/debunking-prince...

Is this being posted for humor value?

This was written in 2014, the researchers saw a small trend and extrapolated that the trend would continue, causing 80% user loss from 2014 to 2017.

Since 2014, FB's active user count has increased from 1 billion to 1.7 billion and the stock price from ~60 to 120.

This article is a great example of a very terrible and very wrong prediction.

There are around 3B internet connected people in the world, does FB really have half of them? I know more than a handful of people that eschew Facebook entirely, it's hard to beleive that half of the world is on FB. I also have 3 or 4 FB accounts myself, (that I created to test an app) am I counted as 3 people?

1.79 billion monthly active users.

I guess you might be counted as 4 people if you use all your test accounts regularly, but the average number of Facebook accounts per person must be extremely close to 1.

Facebook estimates that between 5.5% and 11.2% of accounts are fake [1]. So that's still around 1.59 billion real users.

[1] http://thenextweb.com/facebook/2014/02/03/facebook-estimates...

I wonder how many active Facebook users aren't actually actively using Facebook. Per that link, Facebook defines an active user as anyone who logs into a Facebook account, but that includes using a Facebook account to log into a third-party service. An enormous number of apps and services now use Facebook OAuth rather than their own user registration system and, anecdotally, I know a surprising number of people who have zero-friend zero-post Facebook profiles solely because of that. As far as I can see Facebook doesn't publish information on how many of their active users are actively using Facebook itself.

But Facebook has sort of outgrown Facebook itself.

The value to outsiders is now about the untangleable web they've built: so whether you're using a dating app, OAuth, Messenger, Instagram or Whatsapp (last one is a bit of a tricky one right now) you're still a user of Facebook. Maybe you're no longer addicted to the feed - cuz fake news - but Facebook is still deriving some value from you as you're still a part of their platform.

Facebook is an ad platform on top of a social media network. It's correct to measure those OAuth users as active, because they're still being followed around the web (any site that uses the FB platform, which practically is most), they're still part of the monetization platform, they can be targeted in ads, etc. That's what's really relevant to their business, not how many people post content on their FB wall or send messages.

This is no longer true as of late 2015. The new definition is "We define a monthly active user as a registered Facebook user who logged in and visited Facebook through our website or a mobile device, or used our Messenger app (and is also a registered Facebook user), in the last 30 days as of the date of measurement."

So if they don't use Facebook's website or app, or Messenger as a Facebook user, they don't count.

http://www.adweek.com/socialtimes/monthly-active-users-defin...

Facebook has no idea how many fake accounts exist and neither do I. But it is definitely higher than 11.2%. That estimate was self-serving for business purposes not a serious analysis.

Given sites like these http://www.followersandlikes4u.com/buy-real-facebook-followe... i would put fake accounts at 30%

You're saying that buying fake followers accounts for a 360 million user discrepancy (from 10%-30%)? Only a small subset of app developers, page maintainers, etc, buy fake followers, and most people are not app developers. And FB deletes fake accounts all the time. 6-12% is much more realistic than 30%. Most apps/networks have between 2-10% fake/dupe/invalid accounts, and you become a bigger target the larger you get.

What about Twitter? Do you think they have between 6 and 12% as well?

I would assume you are counted as 3 or 4 people. I think this is restricted by the Facebook ToS and is one of the reasons they try to collect real names, which admittedly doesn't really fix the issue.

Facebook also allows businesses and organizations to have accounts, I'm pretty sure.

The last Pew Research poll shows 79% of "online" US adults use Facebook. That's equivalent to 68% of all US adults, which at least makes it seem plausible that half of all people with access to the internet use Facebook.

http://www.pewinternet.org/2016/11/11/social-media-update-20...

Extrapolating from the US number to people with different languages, cultures, and legal environments all over the world seems like a leap. For instance, Facebook is blocked in China.

But heavily used in other countries. Travelling through southeast Asia and India makes me believe that usage there could be even higher than in US or Europe, nearly everyone who has internet on their phone will use Facebook.

I guess they have nowhere as much, but it's probably good for marketing purposes. In my social environment probably 25-30% of the people are on facebook. And I guess that's still way above the average - because of age (the percentage of > 50 year olds will be much less) and region (in less developed countries less people will have the chance to use facebook).

In total I would expect less than 10% of the world population to use Facebook, but even that would be massive.

It was a demo page on my HN clone ( with tags) and i thought it was an interesting throwback--> http://tagly.azurewebsites.net/Home/Demo

The rapid icarus-like ascent and fall of MySpace should be enough to show people that social media doesn't behave like traditional businesses. Extrapolating out trend lines over the course of three years in a medium that is not quite twenty years old seems to be asking for trouble, in my opinion.

>This was written in 2014, the researchers saw a small trend and extrapolated that the trend would continue, causing 80% user loss from 2014 to 2017.

Do you not see the humor value in that?

A FB engineer's response at the time: https://www.facebook.com/notes/mike-develin/debunking-prince...

Not everyone at FB is an engineer.

Engineer is liberally defined nowadays. It apparently no longer means someone who has a B.S.(or above) in Engineering.

It also no longer means someone who shovels coal into a steam engine

Actually, that was the fireman.

I stand corrected

oh man, that's great. I wish they had left it more snarky and hadn't soft-sold it by throwing in the "air" bit at the end, but that's just me.

Nice article! But sarcasm doesn't work as well when you have to explain it on the end.

Reminds me of doing a math problem, getting an answer, and saying "well the math is correct so this value must be as well" and not actually thinking if the results really make sense.

9 women can make a baby in a one month. I used billions of data points to train my model.

My data model says 9 women and 9 men.

My data model says 9 women and 1 man.

My data model doesn't say all that much in english but has amazing curves that I can integrate with the data I've got. Any intelligent person who sees the result can easily differentiate it from what's out there.

You get more insight into curves on Facebook if you analyse them with the Cox-Zucker machine. No relation to the facebook founder.

its impossible to predict when FB will die (most companies die). its like timing a market top, we know its coming but getting exact timing is hard.

just some anecdotal facts: i know lots of my friends (mid age working professionals) realize that FB is a waste of time and if anything try to avoid it more and more. its just a matter of time before other platform takes over.

Zuckerborg is going to keep this thing around at least another 10 years. I mean I am skeptical of Zuck. But he's got some sort of management genius. So much so that I think it's not a bad idea to elect him President in 10 years.

While I agree he's likely looking into the possibility of a political career, I'm not sure anyone will be elected President in 10 years.

Please explain.

there isn't a presidential election in 10 years.

4 year election cycles, so 10 years from now would be just after midterm elections.

For starters, there's no (planned) presidential election in 10 years. There's one in 4 years, 8 years, and 12 years

But he might be making a prediction/silly joke that Trump will destroy the US or something.

This last presidential race more or less started 2 years ago. If he were to run for President for the election in 12 years, he'd more or less start running in 10.

No I was just going by Zuck's age.

He's 32 years old, and the President has to be at least 35 years old. So he could run in 4 years.

I would not want a 35 year old Zuck as President.

They realize, stop using it for a few weeks/months, starts missing it and then come back again. I myself have gone through a few such cycles. This is frustratingly similar to trying to quit an addiction.

I see a lot of it too .. the fact that a lot of people want to stop using it is telling us whats coming for the company..

True, but I'm envious that my startup isn't in a situation where people were trying to quit it.

I went through several cycles like that. But somehow I stopped using Facebook in 2013. I managed to shift the mindset from "it's a waste of time" to "it's very shameful to use facebook" (and maybe connect it with some really shameful thing in your mind, like, "using Facebook is the same as stealing an old man's money") and that magically works! And it doesn't require any iron will for me.

After that, every single time I think about using Facebook the shame engine in my mind generates enough repel force for me to not do it.

I didn't do that whole process intensionally. I remember seeing some meme like "checking out Facebook in the summer -- all my friends are having fun somewhere and me masturbating lonely in a dark room" and that initiated the mindset change.

To be honest, I know using Facebook is NOT a shame, but now I don't have any reason to use it at all. (There are some stupid website which only allows Facebook login, and I use my gf's account for it -- neither I nor her care about what's on her facebook.)

One trick that worked for me is to gradually subscribe and like boring stuff like cat pictures, scientific journals ect. Anything that doesn't rile you mentally, i.e low mental imprint. Do this very very slowing but consciously after a while your facebook feed would become 'boring' making it easier to quit.

I did this couple of years ago and never went back.

Unless you're looking to wear your non-facebook status as a badge of pride (which many do), I think there's a more constructive way to quit wasting time on FB.

Every time you see a vapid post, unsubscribe from whoever posted it. Eventually you'll have five minutes worth of feed every day and it's only people you're interested in. Suddenly you're not wasting time, you're just getting a couple friend and family updates with your morning coffee.

More radically, you can try this chrome extension. I've been using it for a couple of weeks now and it drastically cut my use of facebook.

Link: https://chrome.google.com/webstore/detail/news-feed-eradicat...

i use that now but it's hard for an addict to install that .

What I did was to simply log out from Facebook.

I would impulsively type the URL and then the small nuisance of having to log in was enough for me to close the tab. After a few days it was enough to stop the compulsion, and now I haven't used Facebook in months.

Reddit on the other hand...

I even deactivate FB. Isn't enough to keep me out for long.

When there's lots of people on your FB, it's mostly noise and very little signal. When there's few (as mine is now), you realize there's still very little signal. I could never get over the fact that, IMO, broadcasting on a social network isn't meaningful. This is the main reason why I really like dedicated one-on-one messaging (ex, WhatsApp) but pretty much find FB proper useless in 9/10 cases (Events being the only nice aspect). That being said, I hold out little hope that FB will finally die.

I unfollowed most people and started liking things. My feed turned into a sort of ghetto Google Reader... but they always tweak it to meet Facebooks objectives.

To be honest, the only content left is the mommy swap pages. Fancy Craigslist. :)

The signal fell off a cliff: when I used it 10 years ago in uni, it was actually useful (and the algorithm did a better job of showing things you wanted to see/there were less ads of all types); now it's mostly a proxy for emails about parties and spotify/hulu/tinder login.

I think Facebook has failed to notice the precipitous drop, because they can only A/B test small changes, and none of the thousand cuts seemed like much damage.

Now they lack the magic to get new users hooked, are transparently a psychologically-manipulative ad-network, and are wondering why engagement is dropping or young users are finding new platforms.

I call it stats blindness: you optimize for a proxy, and end up walking slowly downhill.

I never used Facebook until I ended up as secretary of a sports club. It adds a lot of value for messages such as, "Could someone who regularly drives past the bottle shop please pick up a carton of beer and drop it off next time they come to the club rooms?" If you post that on Facebook, people read it a lot sooner than if it's on some group-specific message board. Then, whoever offers to do it gets to have the whole club see them volunteer.

If you're trying to organise a community instead of just making money from its economy, then a semi-public space like Usenet or Facebook turns out to be really useful.

It's true that most companies die, if you include every tiny startup and shop. I don't think it's necessarily true if you look at companies that are the size of Facebook.

There are a lot of very, very old companies in the world: https://en.wikipedia.org/wiki/List_of_oldest_companies

Kongō Gumi is a Japanese construction company that was started in 578 AD, and operated for over 1,400 years until it was bought in 2006.

If Facebook lasts as long as Kongō Gumi, they'll still be around in the year 3432.

Well, LiveJournal and MySpace are still around, as is AOL. But I think the big question is when Facebook will stop being the "it" page, and I'm guessing sooner rather than later. A lot of people already seem to be rather sick of it, and would jump on a halfway decent alternative if it arose (and targeted a community, rather than just being a "Facebook alternative").

Do you think those people are sick of Facebook or sick of the "connected lifestyle". For me it's the latter, especially when you hear "what, your not on Facebook?"

I've heard people be suspicious of other people that didn't have an online presence. Especially if they were working in tech. It's just so weird sometimes.

I guess it depends on what you mean by the "connected lifestyle." A focus on rapid-fire micro-updates sent out to a general audience instead of a more personalized and thought out conversation? I think there are a number of people tired of that. But alternatives don't have to follow this model. They could be based around longer, more thought out and more personalized correspondence (like e-mail), or on developing real world communities (like Meetup).

I think Slack is a good example of this. What it's doing is possible on other platforms, but it's better at what it does and so it's attracting a lot of attention. Eventually some site like that is going to attract enough people who start using it as there primary online interpersonal interaction page, and let their presence on other sites stagnate. At which point other people will start navigating to the platform simply because that's where everyone's going, and you'll get a snowball effect.

Like with a lot of things, once it gets started it will probably happen very rapidly.

The alternative may be putting down the phones and computers and talking to people face to face again. Said as I stare at my phone on the bus with most other people doing the same.

But... Is using a phone on the bus a problem?

Buses (to many people) are a place better off without a lot of conversations going on. That used to mean books or staring out the window. Phones seem to offer a legitimate improvement here.

Now, if you're talking about somewhere like bars, that might be different. But I mostly see face-in-phone on buses, and people talking at bars. (YMMV, anecdata, etc.)

> A lot of people already seem to be rather sick of it

I suspect any issues they have with Facebook would be had for any of its alternatives.

Unless that alternative is radically different, like in-person conversation.

Firstly: https://www.statista.com/statistics/264810/number-of-monthly... Facebook's MAU is now at 1.7B users (That's almost 1 out of 4 people on the planet!) and is growing at about 10% per year.

Secondly: Saying that Facebook's popularity will rise and fall in a similar trend that Myspace did is like saying Google's popularity would rise and fall like Ask Jeeves's.

Thirdly: Facebook owns Instagram and WhatsApp

Conclusion: Big brother Zuck is here to stay!

So sad that Jeeves is no longer around :(

I think Facebook's demise is closer than we believe, and that the main culprit will be Wall Street.

Given that the Street wants the stock price to always go up, future earnings has to go up. Hard to do with less user growth, because 1.5B already. More ads is hard if you want to keep the platform interesting. Still, room to grow in lesser FB-developed countries, with FB benefiting from economy growth and ad market growing there. But how much growth left in main earning market (US) without destroying the user experience?

With strong leadership, FB could say "to hell with higher and higher profits, I am happy to run a very profitable utility, not plastered with ads, a la Craiglist". But Zuckerberg seems all talk there. And he is trying to sell a lot of his stock. Yes, he's trying to retain control too, but that will be very hard (see current litigation on that). And employees don't like stocks that are not going up.

With less user and financial growth, less momentum, fewer interesting challenges, FB can slowly turn into Yahoo/AOL, and people may start saying bad things about the company (users + employees), creating a bad momentum. It would be hard to reverse, especially if FB is managed quarterly by earnings, as opposed to strong leadership.

Also, I get the feeling that FB got a small temporary break with Trump. A Hillary victory may have meant many very senior executives leaving to the new administration, possibly accelerating FB's loss of momentum.

Obviously, lots of talk here, many "ifs" and "maybe", and no hard data. That's why in the end I am not shorting the stock today. But wondering if I should.

Back in 2008-2010, FB conquered instant messaging [1]. Since then, other services proliferated, including WhatsApp (which they wisely bought), and others, but FB Chat, rebranded as Messenger, is still a hugely important platform.

Instagram is growing among the demographic that Facebook first conquered when it first came out and is vying [2][3] for Snapchat (partially by its original premise, partially by copying its features [4]) to be the most popular platform for 18-25.

Meanwhile the original Facebook's audience is aging, and most of the time it's an event scheduler (like Google Calendar), or photo album store (like Flickr), or a tumble-log of memes (like Tumblr), but lots of people still run the app on their phones and wander around with Location enabled. On desktops, a ton of people don't clear their cookies and the Like widget on websites lets Facebook know what sites they visit. Some people have taken to using Facebook Auth to log in to medium-value sites they visit less often -- like to buy concert tickets.

All this contributes enormous value to Facebook's ad ecosystem, which is in second place to Google, and together those two are the only real players [5].

[1] https://news.ycombinator.com/item?id=11107321#11114518 [2] https://news.ycombinator.com/item?id=12799794#12800418 [3] https://news.ycombinator.com/item?id=12799794#12800957 [4] https://news.ycombinator.com/item?id=12210324 [5] https://twitter.com/jason_kint/status/814842452003659776

I read the linked paper and it's of a very poor quality. They fit an epidemiological model to the current data and then extrapolate it blindly to the future. (The fact that it worked in the case of MySpace is not as impressive as it sounds since by 2014 MySpace was long dead.)

I'm almost two years FB free but I don't see anyone else quitting, just allowing it to suck them in 24/7. And new users are always joining.

I uninstalled fb app and messenger but thats the most i can do. I can't quit facebook because my social life depends on it. I have 3-4 group chats on facebook, events i go to are announced and promoted on fb and as someone who went to a different country for Uni, it's a great way to be in touch with friends from high school occasionally.

I've also been off for quite some time now (I "deleted" my account 3 years ago). I know none of the information gets removed so I logged back on early last year to grab some old pictures for a wedding slideshow. It's really scary how much I wanted to keep it activated after re-activating it, and how quickly people started messaging me about why I was back on Facebook. I'd say it's less like the plague and more like gangrene.

This made me go back and find a couple of Facebook threads right after IPO that made predictions:

https://news.ycombinator.com/item?id=1719975 (Facebook is not worth $33 billion) - This was more than a year before the IPO. The market cap is 10x that today. Lots of poor predictions but also some cautious optimism from some people.

https://news.ycombinator.com/item?id=4002730 (FB down 8% at opening) - Some rubbish number pattern matching and lots of doom and gloom. Interesting

I found myself spending/wasting time with fb. I tried to close the account couple times but few days later reopened the account.

Last year, I tried to close the account again. First unsubscribed from all email notifications from fb. And downloaded all my content from fb. Then closed the account. But this time too I logged in twice but made sure instead of logout I closed the account both times. That's it - its been more than 8months now. I didn't bother to login to fb. Honestly, I feel much peaceful now.

Last week decided to quit, en.lichess.org its been week . let's see :-)

Try Sublevel[0] if you feel a social itch.

0.https://sublevel.net/

"John Cannarella and Joshua Spechler ... have based their prediction on the number of times Facebook is typed into Google as a search term. The charts produced by the Google Trends service show Facebook searches peaked in December 2012 and have since begun to trail off" as more users started using their phone instead of google to experience the Facebook.

I think that this may be true in that the desktop app has considerably lost users in favor of the mobile app.

Search traffic has decreased significantly:

https://www.google.ca/trends/explore?date=all&q=facebook,yah...

I had a dream recently (after the election that shall not be named) that Facebook's stock was spiking like an airplane about to stall. Indeed, it did stall and crashed like a rock.

It feels to me like there is a reckoning coming that even Facebook is not immune to.

Based on Google Trends, which is the source used in the paper, Facebook is only 43% as trendy now as it was at its peak in 2012. Compared to the past, Facebook is as trendy now as it was in 2009.

The irSIR model that was predicted by the paper is about 2 years off from where we are now (the "late" model is about 1.5 years off). Repeating the study with information up to today and the new prediction is that Facebook's Google Trends measure will drop below 20% within 2 years.

https://www.google.com/trends/explore?date=all&q=%2Fm%2F02y1...

A great example of how most headline-grabbing "studies" are garbage.

Ever tried to quit facebook? What happened?

It's kind of like Hotel California.

I quit a while back. I don't miss it.

The closest thing to a problem has been that I sometimes learn about events a little late. But I've found that pretty much anything I'd actually want to go to I find out about other ways, usually by talking with friends. I probably miss out on a lot of major life news from people I vaguely know, but I'm okay with that. And it probably pushes me to make a little more effort to stay in touch with people in old-fashioned ways, like seeing them in person, or talking on Snapchat.

So far, I've had a somewhat similar experience. I am more energized to talk to people individually on Signal, WhatsApp, Facebook Messenger, etc. because of an interest in staying up to date on their life events, which personally feels healthier to me.

I deleted Facebook about 2 months ago. Nothing bad happened. In fact, I feel relieved more than anything that I don't have this data-collecting overlord following me around. Also it takes out the stress of possibly "missing" Facebook messages or invitations.

I can't use one or two dating apps anymore, but I have a feeling my relationship status is unaffected regardless.

Contemporary discussion: https://news.ycombinator.com/item?id=7104904

I think the most interesting thing about this is the previous HN discussions[1][2].

It's worth noting that even back then people thought it wasn't a great piece of research.

[1] https://news.ycombinator.com/item?id=9096125

[2] https://news.ycombinator.com/item?id=7104904

Yeah, I totally didn't believe it then either. I wouldn't doubt they'll lose users one day, but I seriously doubt they'll lose many anytime soon.

The title is a bit clickbait-y. From what I gathered, they really mean 80% of active users which is probably still too high, but I think it's probably Snapchat's year to peel off A LOT of active users from Facebook. Not that it's a zero-sum game, but I anecdotally I see a lot of FB fatigue around me.

Yep. FB fatigue is REAL!

Whenever I see a prediction from academics, I'm reminded of this: http://www.msnbc.com/the-last-word/watch/scientist-predicts-...

"Princeton professor predicts 99% chance of a Hillary Clinton win"

This is a little unfair, it's not clear to me why academics would always have worse predications than any other body of people. This especially goes for the presidential outcomes predictions.

Especially when most predictions were for her.

Also, if that summary is true, that meant 1% likelihood for Trump. Not zero. That's how stats work.

> it's not clear to me why academics would always have worse predications than any other body of people

You're right; it's a problem with predictions _generally_.

It's just that academics make predictions that carry more credibility.

"Assuming a perfectly spherical voter..."

Whenever I comments like this I like to ask what they think the chances of losing the popular vote by 2.8 million (or whatever it ended up as) and winning the electoral vote actually are?

Maybe she did have a 99% chance of winning? I don't think so - but I haven't actually worked it out. I'd be interested to see the conditional probabilities.

538 put Trumps chances at ~20%, which seemed about right overall.

Facebook still may lose 80% of users in the near future. AOL Instant Messenger used to be the shit and everybody used it...

Have been using the Facebook eradicator plugin for a long time. No news feed, just the groups and messages. Perfect for me.

I've long stuck to just groups and messages over a web browser. But this extension can surely help people who just keep scrolling down the news feed and spend several hours a day on it without realizing. I found a similar extension for Firefox called Kill News Feed for Firefox [1], but it was last updated in 2014. I'll give it a try sometime.

[1]: https://addons.mozilla.org/en-us/firefox/addon/kill-news-fee...

The article says "80% of its peak user base within the next three years", but the title says "by 2017"?

Also, I'd have liked to know to what Facebook would lose its users. During the Pokémon GO craze Facebook and other social sites lost temporally their users before they came back.

"John Cannarella and Joshua Spechler, from the US university's mechanical and aerospace engineering department, have based their prediction on the number of times Facebook is typed into Google as a search term"

Really amusing. Now people type facebook.com and open facebook app on mobile. This research is garbage.

"based their prediction on the number of times Facebook is typed into Google as a search term"

While it might have been common to go to a website by searching for it, accidentally or not, many interact with Facebook via a mobile app. Using google search trends will not account for this change in behavior.

> most companies die

I feel like Facebook and Google will never die, at least not during my life. Am I the only one?

AOL and Yahoo aren't dead. Nor MySpace.

The real question is if Facebook and Google will always be incredibly popular.

No, I am sure are not the only one. This is a temporary perception and it could change with just a single event.

Take VW (Volkswagen) for instance: They looked like the dominant car vendor, but with the Dieselgate they hit a wall which will make it hard to recover. Another example would be Nokia. They still sold about 40% of the phones in 2010 - but they missed the crucial trends and made the wrong decisions.

How do you think about IBM, Intel and Microsoft? 15-25 years ago people had these feelings about these companies. And many people from back then are still alive.

PS: I personally would believe Intel lives longer than Facebook.

Facebook are smart and moving with the times - no doubt the original Fb blue website is losing hits etc but they move on with Instagram and Messenger which are more popular than ever - Insta I think is going through an intense growth phase at the moment

These things seem to come in cycles - as one service begins and attacts users the other one declines. We are not seeing the alternative to Facebook yet. In other words users are not going someplace else...yet, but they probably will.

The alternative to Facebook already exists: Instagram.

That's why it it was such an inspired move by Zuckerberg to acquire it when they did.

Hmm. and I was just wondering how much longer Google would last. How often do people really use Google? Most everyone I see is just taking & sharing photos / occasionally updating a facebook page.

Google is the #1 source of knowledge for most internet users in the world (excluding China). In addition to that they provide a large share of communication (GMail), orientation (Maps) and are responsible for ads on many other pages.

Many companies (including Facebook) tried to beat Google's search algorithm, they all failed (except for Bing which seems to be used by some). Unless someone can make a search engine that is much better than Google, they will keep being successful.

Same for an ad network. If you have a website, using Google for ads seems to be the best option so far. No one is really happy with Google, but they still seem to be better than most competitors.

> Many companies (including Facebook) tried to beat Google's search algorithm, they all failed

I'm not sure how much Facebook tried, but searching within the Facebook platform is a really bad experience. It's as if many things are either not indexed or are indexed with only selected sentences or words from the content. I struggle so many times to find older content for reference and to point people to, and fail a lot more times than I succeed (I mostly search in groups). Since Facebook thrives on people wanting new content and clicks, it doesn't seem like it's in Facebook's interest to make searching within the platform better. From whatever I've seen over the last several years, if at all Facebook has been working on search, the engineering team is utterly incompetent on that front.

I remember that there was this huge announcement about Facebook Graph, which was Facebooks idea of a revolutionary new way of searching. Didn't work at all, and searching on Facebook is indeed very limited. Which is somehow really problematic, as Google also cannot really search Facebook.

Wikipedia with some background: https://en.wikipedia.org/wiki/Facebook_Graph_Search

Can FB really be compared to these diseases though? Maybe I'm nitpicking the details here but a lot of diseases, including bubonic plague mentioned in the article, kill a large proportion of their hosts.

I'm a privacy nut, and to me FB is worse than these diseases and will kill society, progress, freedom from surveillance and many other things that will deny more people the rights that they ought to have without being oppressed or suppressed.

Phew!!! What a searchers. It just means fb is as or more popular than google. You dont need to go to princeton to guess that. The funny part is, you don't even need to read the article to make it out.

I recently quit Facebook, but I get the impression that I'm the exception.

I quite FB in around 2009 with the plan of being more productive and it worked really well. I was being constantly distracted by unimportant events from other people. The one interesting thing that happened after that is that some of my "friends" actually contacted me to see if everything was ok with me, when in reality they where validating their egos making sure I didn't block them.

I didn't find my productivity went up. Evidently, I can find enumerable things to distract myself with. I just got fed up with all the political hate flying around. Seems this year things went from people ironically commenting on the things our government does to claiming that anyone in disagreement belongs in a death camp. I just didn't want to see that shit anymore.

I cannot enumerate the things I can distract myself with; I can therefore find innumerable things to distract myself with.

More productive

Attending fewer events

Misreading others' intentions

Yeah, I think I'm gonna stick with Facebook. I don't need even MORE reasons to be a hermit -- it's already my natural tendency!

Thats fine if it works for you. To me its just noise that pollutes all the things I want to do. You don't have to be a hermit. If avoiding Facebook makes you a hermit, than I think we failed and we're depending too much on a single technology to be social, that doesn't sound right to me.

I chose not to sign up immediately after witnessing what it did to my friends. I feel I did right thing. a month ago, my grandma who lives in a very remote place (in rural India) was dissapointed to not find my profile or pic in fb app.

This really points out the danger of using an analogy to extrapolate other similarities. Bacteria don't live forever, but there's nothing to prevent Facebook users from being Facebook users forever.

Facebook users also don't live forever, to be fair.

They live forever in Internet years. At least most of them do.

a lot of people have an app on their phone. that means no search.

They've already lost 80% of me.

Maybe, but what if people just open the app now and don't search google for facebook?

What a coincidence Snapchat is supposed to IPO in 2017...

$58 in 2014, $120 today? Follow the money.

There may be a Facebook app to blame.

This is proof that the signal-to-noise ratio of top links on HN is declining.

https://xkcd.com/605/

It is 2017

The article is from 2014.

vkontakte fberyoad, bitches.

The comparison of Facebook to a disease is entertaining (if because it has a little bit of truth from a certain perspective). But I'm not sure that the model they chose is the right one to use to make predictions. So I think immediately the predictions have to be in question.

That said, I would like Facebook to go, and not only because of its repugnant behavior with regard to its partnership with the US and foreign governments for propaganda, censorship and surveillance.

About a decade ago developers tried exhaustively to make every app "social". The reason for doing so is that, after a certain threshold, having a community locked into your platform raises barriers to competition.

When it comes to large social media sites - Facebook now and Myspace before it - there can 'only be one'. There's only one place where large numbers of users can go. They don't want to post everything they think ten times to ten different communities.

This makes the kind of large scale social media network Facebook represents a "natural monopoly". It can not have competition, except in the form of potential successors to market capture.

This makes the "market" of social media look more like king-of-the-hill than it does efficient laissez faire competition.

The better solution for social media would be a distributed platform with no central ownership. This solves privacy, censorship, and surveillance concerns and limits propaganda to a certain degree as well.

It also creates an environment where there can be competition. Now companies are free to create different client software with different feature sets at different costs. So distributing the infrastructure for social media addresses the enormous problem of market failure as well.

This makes no sense. Natural monopolies arise when there are extremely high fixed costs of distribution, like Utilities, Railways, things that are difficult to build. That is not what social media is at all, and there is absolutely nothing guaranteeing that what those Princeton researches predicted will not become true, precisely because software on servers is so easy to replicate, and distribution is inherently word of mouth. Facebook had to purchase Instagram because they couldn't figure out that type of social media. They may not be able to do that for the next Instagram. Facebook has started to suck hard because they inherently live based upon outrage and fake news. If they can't figure out ways to constantly reverse various trends that come up over time, they could get caught in a crap information spiral and fail.

I think you might find that the fixed cost of distributing pictures and video to 1+ billion users who check their FB hourly to be rather expensive. I think you would also find that storing all those pictures and video might be expensive. Additionally, the database requirements to search through those billions and billions of posts are not going to be small and on top of that, FB's database returns the results very quickly.

Sure, distributing the software to new servers is free. Procuring the software that needs to be on the server is another matter. This isn't a slap a load balancer on top of a few servers and a Postgres instance sort of problem.

It's not just, "I think you might find," but rather, that is precisely correct - server costs for 1 billion users is expensive. However, do you need to set up for 1 billion users in year one? Which is more expensive to start up: a railway or an internet-based software application with a very high growth rate based upon word-of-mouth?

"because software on servers is so easy to replicate, and distribution is inherently word of mouth. "

Networking effects are extremely difficult to break.

Also - brand power is stronger than you'd imagine.

Coca Cola is not better than many other colas. Taste tests confirm this.

Yet coke and pepsi have a 'brand duopoly' on colas.

Cola should be a commodity, like water, but no, it's a weird duopoly/oligarchy.

Aside from brand, and networking effects - there is usage history etc. - Facebook also has all of your social history.

Facebook does exhibit many monopolistic advantages, as does Google.

Facebook started in small social groups (Universities) and grew from there. I'm sure that if someone comes up with a better social network, it will quickly be adapted. Snapchat shows that you can become extremely popular with a simple idea that people like.

But Facebook is at the moment really good at capturing what users want. So they have this monopoly for a reason. But I'm very sure that at some point people will leave Facebook for something else, just because that's how it has always been. The question is just how long it takes until that happens.

> But I'm very sure that at some point people will leave Facebook for something else, just because that's how it has always been. The question is just how long it takes until that happens.

Though I'd wish for this to be true, I'm not sure it'd happen anytime soon. Gmail, though not a social platform by itself, stands as an example of an entrenched email provider on the consumer side because of its features (and originally a promise of larger and increasing storage quota). Similarly, Facebook is not sitting on its monthly average users and will keep improving, acquiring and absorbing things from other platforms to keep its stronghold. Sure, it will face challenges (like Snapchat), but it will not be dethroned soon.

I don't think many people saw the decline of Yahoo or MySpace coming.

Companies can very quickly become unpopular. They will exist for another decade or so before being bought for a cheap price. I'm very certain it will happen to Google and Facebook just because so few companies dominate a market for more than 20 years or so, especially if there are no large start up costs.

However, that doesn't mean one company will take over what Google or Facebook do. It could as well be companies doing something else, but taking away the main revenue streams.

Fake news!!

FB usage has shifted.

I wonder how many people use it in the very social sense anymore, and how many use it for basic messaging.

Reply on news.ycombinator.com