> 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).
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.
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).
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?
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.
> 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.
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:
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.
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.
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?
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.
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.
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.
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.
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.
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.
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.
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.
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].
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'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.
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.
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.
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 :-)
"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 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.
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.
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.
[+] [-] ysavir|9 years ago|reply
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).
[+] [-] azinman2|9 years ago|reply
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.
[+] [-] cag_ii|9 years ago|reply
[+] [-] logicallee|9 years ago|reply
[+] [-] throwaway7645|9 years ago|reply
[+] [-] abiox|9 years ago|reply
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.
[+] [-] dools|9 years ago|reply
https://en.wikipedia.org/wiki/Timeline_of_Facebook#Product_a...
[+] [-] StringyBob|9 years ago|reply
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.
[+] [-] ZeljkoS|9 years ago|reply
[+] [-] NicoJuicy|9 years ago|reply
[+] [-] sjg007|9 years ago|reply
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.
[+] [-] firebones|9 years ago|reply
[+] [-] paulhodge|9 years ago|reply
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.
[+] [-] Johnny555|9 years ago|reply
[+] [-] vacri|9 years ago|reply
[+] [-] NicoJuicy|9 years ago|reply
[+] [-] kerbalspacepro|9 years ago|reply
Do you not see the humor value in that?
[+] [-] robobenjie|9 years ago|reply
[+] [-] rconti|9 years ago|reply
[+] [-] gohrt|9 years ago|reply
[+] [-] kresimirus|9 years ago|reply
[+] [-] cblock811|9 years ago|reply
[+] [-] laxatives|9 years ago|reply
[+] [-] readhn|9 years ago|reply
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.
[+] [-] personlurking|9 years ago|reply
[+] [-] maverick_iceman|9 years ago|reply
[+] [-] nathan_f77|9 years ago|reply
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.
[+] [-] devoply|9 years ago|reply
[+] [-] thatguy99|9 years ago|reply
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!
[+] [-] paulddraper|9 years ago|reply
[+] [-] quickConclusion|9 years ago|reply
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.
[+] [-] niftich|9 years ago|reply
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
[+] [-] maverick_iceman|9 years ago|reply
[+] [-] bluedino|9 years ago|reply
[+] [-] mattnumbe|9 years ago|reply
[+] [-] parthdesai|9 years ago|reply
[+] [-] arjie|9 years ago|reply
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
[+] [-] giis|9 years ago|reply
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 :-)
[+] [-] curiousgal|9 years ago|reply
0.https://sublevel.net/
[+] [-] dmalvarado|9 years ago|reply
[+] [-] bhouston|9 years ago|reply
Search traffic has decreased significantly:
https://www.google.ca/trends/explore?date=all&q=facebook,yah...
[+] [-] blurbleblurble|9 years ago|reply
It feels to me like there is a reckoning coming that even Facebook is not immune to.
[+] [-] syphilis2|9 years ago|reply
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...
[+] [-] pfarnsworth|9 years ago|reply
[+] [-] slantaclaus|9 years ago|reply
[+] [-] skewart|9 years ago|reply
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.
[+] [-] rubyfan|9 years ago|reply
[+] [-] corvallis|9 years ago|reply
I can't use one or two dating apps anymore, but I have a feeling my relationship status is unaffected regardless.
[+] [-] pessimizer|9 years ago|reply
[+] [-] nl|9 years ago|reply
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