10 ms·

What Happened When Facebook Hired Some Journalists

102 points by hollaur 10y ago

andrewfromx 10y ago
"After doing a tour in Facebook’s news trenches, almost all of them came to believe that they were there not to work, but to serve as training modules for Facebook’s algorithm." ha! aren't we all.
- mc32 10y ago
  So.... They finally know what it feels like to be a blue collar worker whose job gets outsourced and sometimes are offered benefits contingent on training their offshore trainees. Everyone was happy to have $30 jeans which could last one year over their $70 jeans made stateside which could last five years or more... But now that it affects them, oh my, this is bad now...
  - Retra 10y ago
    >Everyone was happy to have $30 jeans which could last one year over their $70 jeans made stateside which could last five years or more... I'm not sure this analogy works. Surely the people selling jeans would prefer to sell new $30 pairs every year rather than $70 pairs every five.
    - GFK_of_xmaspast 10y ago
      What if that $70 jean costs them $60 and the $30 jean costs them $29.
    - mc32 10y ago
      Of course. Buy buyers mainly see "cheap jeans!". They don't always think about quality, craftsmanship, etc. They've been primed for yearly changing fashion which enables cheap jeans made from sub par materials.
      - slavik81 10y ago
        Even if they did care, it's not easy to tell until the jeans start falling apart.
        B1FF_PSUVM 10y ago
        Don't they come that way now, new from the store, all torn up and stuff?
    - majormajor 10y ago
      A lot of buyers would too. Overall price premium, but lower immediate costs to keep up with the latest fashions. See: H&M, etc.
    - EGreg 10y ago
      I see this as a fundamental problem of the unsustainability of our economy. The incentive of every corporation is to earn an increasing amount of money, and thus to move more things from resources to garbage as quickly as possible. "But all those jobs!!" is a stupid argument. Jobs aren't the goal. The goal should be getting resources to the right people, and sustainability for the long term.
      - harmegido 10y ago
        I don't buy this planned obsolescence bs. I get that most markets aren't very free, but they're free enough to ensure that superior products generally win out. Companies are trying to make money today. They aren't thinking: "hey, let's put out a crappy product that will be worse than our competitors' so that people will have to buy more of our stuff in the future!" Any company that did that wouldn't make it to tomorrow.
        EGreg 10y ago
        "Superior" is a subjective term. Superior in what sense? Trump for example got tons of coverage. Is he "superior" to someone who may be more knowledgeable, skilled and would "make a better president" but got less coverage? The amount of advertisement, exposure, and just plain psychological manipulation matters. Is sugary soda superior to healthy drinks? Is bottled water superior to tap water? So we as a society really need to throw away metric tons of non biogradable materials every minute? Thinking that something wins out because it is "superior" is just confirming the free market ideology - whatever wins must be "superior" is a tautology which makes the word "superior" lose all meaning. Yes, lots of crappy products beat products that performed better in practice, because of external and network effects. But what we should be asking is about the correlation between a company's incentives and polluting the planet. In the main, they are pretty aligned, which is scary.
        quanticle 10y ago
        >they're free enough to ensure that superior products generally win out. Doesn't this imply that the customer has enough information to determine which is the superior product? In reality, it's quite easy to dress up an inferior product with lots of marketing, and gain great success in the marketplace. Heck, look at Beats headphones. They're not superior in any way. Yet, by carefully crafting a marketing message, and putting massive amounts of money behind it, they've managed to get a fairly large share of the market.
  - paganel 10y ago
    > Everyone was happy to have $30 jeans which could last one year over their $70 jeans made stateside which could last five years or more... My only piece of designer clothes was a $160 pair of jeans that lasted for only 3 or so years. I concluded that from now on I should stick with buying $50 jeans which last me at least 2 years.
andrewfromx 10y ago
“It was degrading as a human being,” said another. “We weren’t treated as individuals." hmmm. Trying hard to feel sympathy/emapthy for these victums but my algorithm is throwing an exception.
notliketherest 10y ago
When the employees of Facebook collectively ask "What responsibility do we have to stop the election of Donald Trump" at an internal meeting, it gives me Orwellian chills. Today, it's not governments we need to fear the most, it's data hungry, fascist internet corporations.
- avivo 10y ago
  What if the employees at Facebook instead asked "What ethical responsibility do we have to ensure that we, as the primary distributor of news in the world, do some minimal amount of fact checking." The goal shouldn't be to "stop Trump" — that's not what we want a news distributor to do. I don't want bias against a particular person in my news distributors. What I want is for my news distributors to actually have ethical standards that enable them to serve the public good (and as a by product hopefully prevent lying demagogue's from gaining power). Is it "bias" to use a standardized method or algorithm (even one that involves humans reading natural language") to filter content by accuracy and provide some interface indication not just for phishing, and for porn, but also for falsehoods?
  - morgante 10y ago
    Realistically, this is exactly how they would go about trying to stop Drumpf. Plausible deniability and all that jazz. In fact, decisions which Facebook makes this election season will impact the result regardless of the intentionality behind those decisions. Even something as mundane as the mix between links and images impacts different candidates disparately.
    - makomk 10y ago
      Realistically, the only way for anyone not to seem like they're going out their way to attack Trump during this election is to completely forget about fact checking, because he lies so repeatedly and shamelessly about things he's already been caught out on: https://www.washingtonpost.com/news/fact-checker/wp/2016/03/22/all-of-donald-trumps-four-pinocchio-ratings-in-one-place/ https://www.washingtonpost.com/news/fact-checker/wp/2016/03/...
kristianc 10y ago
> “It was degrading as a human being,” said another. “We weren’t treated as individuals. We were treated in this robot way.” Strange that people at Facebook would feel apathetic toward individuals that can't code.
- drdeadringer 10y ago
  Strange that their job was not to code but to do journalistic duties as journalists.
- dredmorbius 10y ago
  Google has this disease quite badly, from multiple sources. It's pretty common elsewhere in SV as well. And is a cardinal error.
morgante 10y ago
Having worked with journalists, these sound like typical entitlement complaints. Frankly, a lot of writers have an attitude that they're "artists" who shouldn't be rushed and don't have performance requirements. Frankly, it does not sound very hard at all. They have to write 20 posts a day, but each post is only a headline and a brief summary. A focused writer can finish that in 15 minutes. > We had to write in the most passive tense possible. That’s why you’d see headlines that appear in an alien-esque, passive language. Oh no! How dare Facebook strive to be neutral and passive. > After doing a tour in Facebook’s news trenches, almost all of them came to believe that they were there not to work, but to serve as training modules for Facebook’s algorithm Of course they were. Facebook is quite explicitly trying to apply ML throughout their site. Why should they permanently be in the business of employing writers to do something which computers could do reliably and effectively?
- SOLAR_FIELDS 10y ago
  The article also attempts to paint them as overworked and abused and treated differently because they were contractors. Having worked in the big tech industry as a contractor, their description of contractor treatment and workload is pretty normal (no employee perks, quotas, etc.). This is due more to Vizcaino vs. Microsoft [1] than any sort of abusive contractor agenda. 1. http://scholarship.law.berkeley.edu/cgi/viewcontent.cgi?article=1185&context=btlj http://scholarship.law.berkeley.edu/cgi/viewcontent.cgi?arti...
  - morgante 10y ago
    Of course contractors are treated differently. If they were treated identically to employees, the IRS would probably have some questions. They knew they were signing up for contractor position. Frankly, they were lucky to get any benefits at all. The article fails to provide any evidence of abuse though. The worst that Facebook did was not inviting them to company happy hours.
- kyle8 10y ago
  I shall just leave a link to Douglas Rushkoff here https://www.youtube.com/watch?v=87TSoqnZass https://www.youtube.com/watch?v=87TSoqnZass to explain to you the deep flaws in the "default" techie thinking that your comment is a classic example off.
AndrewKemendo 10y ago
almost all of them came to believe that they were there not to work, but to serve as training modules for Facebook’s algorithm If Facebook is doing it's job and utilizing their biggest talent acquisition of Yann LeCun, then this is exactly what they should be doing. A critical part of the path towards AGI is using humans to teach it.
- germinalphrase 10y ago
  Thoughts about the utility of schools to serve this function? We are already teaching humans very simple to very advanced tasks and we do so millions of times a day. Is the fact that tasks/knowledge are being structured for paced human learning irrelevant for this purpose?
  - aab0 10y ago
    It's not irrelevant, since there is a minor area of ML called 'curriculum learning' which asks how to order examples to teach a ML algorithm most efficiently, and it comes up in some other contexts (for example, a boosting algorithm which focuses on optimizing performance on hard/misclassified cases can be seen as somewhat like curriculum learning, and there are variants of gradient descent which focus on hard examples rather than wasting time on cases where the NN can already get the right answer; and for 'active learning', you want to pick the example which will teach the algorithm the most), but there's not much you can take from known pedagogy at the moment and apply straight to NNs. Even the non-bullshit parts of education like spaced repetition have no clear analogues for tasks like 'train an RNN to write news headlines based an article text using this corpus of human-written headlines'.
  - AndrewKemendo 10y ago
    In fact, "expert systems" were built around this principle, though generally more in the industrial context. I don't think that the way knowledge transfer is structured for humans is by default distinct from how we would do it for AI in the long term, but that's a pretty wide range of pedagogy so some tasks make more sense than others for the state of the art today. In fact the way we train ANN/CNN is similar to how infants & toddlers do early classification. Frank Guerin at the university of Aberdeen has done some study in this area that is in it's early stages. As far as I know there have not been efforts to set up schools to train AI's from the general population. The closest thing to it are the efforts by Google and Facebook to train their classifiers (facial recognition, text detection etc...). We are working on an internal process that uses user input to classify objects in the home (furnishings mostly) for auto segmentation and detection.
zxv 10y ago
Reducing the number of journalists implies that supervised learning[0] was effective enough to begin to replace some human effort. [0] https://en.wikipedia.org/wiki/Supervised_learning https://en.wikipedia.org/wiki/Supervised_learning
morning_star 10y ago
why am i reading this this is the most worthless piece of thing i've ever read in my life
- dang 10y ago
  We've banned this account for repeatedly violating the HN guidelines.
  - fallenshell 10y ago
    bless m8
smelendez 10y ago
Maybe Facebook should have advertised this as a temporary "Robot Journalist Training Fellowship"
jccalhoun 10y ago
why would they hire journalists to do this? It doesn't seem like it takes much skill at all to pick a story and write a short summary of it.
- jonathankoren 10y ago
  I wonder that myself. I think it has more to do with standard corporate over thinking about what's skills are actually required for a job. It certainly make it easier to say, "We want someone to write titles and snippets of news stories. Who does that? Journalists. Cool, let's get some of those." disclaimer: i worked on that fb product.
- n72 10y ago
  My guess is they needed the best seed they could generate for their ML algos.
n72 10y ago
'Mark Zuckerberg has been transparent about his goal to monopolize digital news distribution. “When news is as fast as everything else on Facebook, people will naturally read a lot more news,”' Oh lord. That's not news. Those are sensationalist, click baity headlines making us all collectively dumber. It's anecdotal, but I'm quite sure the 'news' my girlfriend has gotten from FB has made her less informed. I actually wrote an FF plugin to hide that right bar, since I would now and then get sucked in by that crap and waste 15 minutes reading about Lindsey Lohan or whatever.
- morgante 10y ago
  Actually, I find that the Facebook headlines are much less clickbaity than the predominant headlines from news organizations.
kriro 10y ago
I don't think the assessment that they were part of a future algorithm is too far off. Given enough expert input picking a fitting image/video or headline should be doable. Worst case you reduce the number of humans needed to one quality assurance person. Algorithm says "this headline, this image"...yes/no;fix.
golergka 10y ago
So, I got interested in this article's description of horrid working conditions and decided to read about it carefully. But I noticed that it gives emotional descriptions of it long before the actual facts. I wouldn't go as far as calling this a manipulation, but it's certainly a disturbing writing style. 3rd paragraph: > grueling work conditions, humiliating treatment, and a secretive, imperious culture in which they were treated as disposable outsiders. 6th paragraph: > “It was degrading as a human being,” said another. “We weren’t treated as individuals. We were treated in this robot way.” And then, finally, on 10th paragraphs, we get a glimpse on the facts: > they received benefits including limited medical insurance, paid time off after 6 months and transit reimbursement (BTW, is it usual for contractors to receive such perks?) > A company happy hour would happen at 8 p.m., and we’d be working Horrible, inhumane treatment indeed. > Over time, the work became increasingly demanding, and Facebook’s trending news team started to look more and more like the worst stereotypes of a digital media content farm. Managers gave curators aggressive quotas for how many summaries and headlines to write, and timed how long it took curators to write a post. The general standard was 20 posts a day. 20 posts during 8 hour work day is almost half an hour on one post. It is considered too little? Seriously? So — apart from all the pretty words, I didn't really see any especially bad treatment. Hell, I'm pretty sure that your average newspaper employees have more nightmare stories.

102 points by hollaur 10y ago

"After doing a tour in Facebook’s news trenches, almost all of them came to believe that they were there not to work, but to serve as training modules for Facebook’s algorithm." ha! aren't we all.

mc32 10y ago
So.... They finally know what it feels like to be a blue collar worker whose job gets outsourced and sometimes are offered benefits contingent on training their offshore trainees. Everyone was happy to have $30 jeans which could last one year over their $70 jeans made stateside which could last five years or more... But now that it affects them, oh my, this is bad now...
- Retra 10y ago
  >Everyone was happy to have $30 jeans which could last one year over their $70 jeans made stateside which could last five years or more... I'm not sure this analogy works. Surely the people selling jeans would prefer to sell new $30 pairs every year rather than $70 pairs every five.
  - GFK_of_xmaspast 10y ago
    What if that $70 jean costs them $60 and the $30 jean costs them $29.
  - mc32 10y ago
    Of course. Buy buyers mainly see "cheap jeans!". They don't always think about quality, craftsmanship, etc. They've been primed for yearly changing fashion which enables cheap jeans made from sub par materials.
    - slavik81 10y ago
      Even if they did care, it's not easy to tell until the jeans start falling apart.
      - B1FF_PSUVM 10y ago
        Don't they come that way now, new from the store, all torn up and stuff?
  - majormajor 10y ago
    A lot of buyers would too. Overall price premium, but lower immediate costs to keep up with the latest fashions. See: H&M, etc.
  - EGreg 10y ago
    I see this as a fundamental problem of the unsustainability of our economy. The incentive of every corporation is to earn an increasing amount of money, and thus to move more things from resources to garbage as quickly as possible. "But all those jobs!!" is a stupid argument. Jobs aren't the goal. The goal should be getting resources to the right people, and sustainability for the long term.
    - harmegido 10y ago
      I don't buy this planned obsolescence bs. I get that most markets aren't very free, but they're free enough to ensure that superior products generally win out. Companies are trying to make money today. They aren't thinking: "hey, let's put out a crappy product that will be worse than our competitors' so that people will have to buy more of our stuff in the future!" Any company that did that wouldn't make it to tomorrow.
      - EGreg 10y ago
        "Superior" is a subjective term. Superior in what sense? Trump for example got tons of coverage. Is he "superior" to someone who may be more knowledgeable, skilled and would "make a better president" but got less coverage? The amount of advertisement, exposure, and just plain psychological manipulation matters. Is sugary soda superior to healthy drinks? Is bottled water superior to tap water? So we as a society really need to throw away metric tons of non biogradable materials every minute? Thinking that something wins out because it is "superior" is just confirming the free market ideology - whatever wins must be "superior" is a tautology which makes the word "superior" lose all meaning. Yes, lots of crappy products beat products that performed better in practice, because of external and network effects. But what we should be asking is about the correlation between a company's incentives and polluting the planet. In the main, they are pretty aligned, which is scary.
      - quanticle 10y ago
        >they're free enough to ensure that superior products generally win out. Doesn't this imply that the customer has enough information to determine which is the superior product? In reality, it's quite easy to dress up an inferior product with lots of marketing, and gain great success in the marketplace. Heck, look at Beats headphones. They're not superior in any way. Yet, by carefully crafting a marketing message, and putting massive amounts of money behind it, they've managed to get a fairly large share of the market.
- paganel 10y ago
  > Everyone was happy to have $30 jeans which could last one year over their $70 jeans made stateside which could last five years or more... My only piece of designer clothes was a $160 pair of jeans that lasted for only 3 or so years. I concluded that from now on I should stick with buying $50 jeans which last me at least 2 years.

andrewfromx 10y ago

“It was degrading as a human being,” said another. “We weren’t treated as individuals." hmmm. Trying hard to feel sympathy/emapthy for these victums but my algorithm is throwing an exception.

notliketherest 10y ago

When the employees of Facebook collectively ask "What responsibility do we have to stop the election of Donald Trump" at an internal meeting, it gives me Orwellian chills. Today, it's not governments we need to fear the most, it's data hungry, fascist internet corporations.

avivo 10y ago
What if the employees at Facebook instead asked "What ethical responsibility do we have to ensure that we, as the primary distributor of news in the world, do some minimal amount of fact checking." The goal shouldn't be to "stop Trump" — that's not what we want a news distributor to do. I don't want bias against a particular person in my news distributors. What I want is for my news distributors to actually have ethical standards that enable them to serve the public good (and as a by product hopefully prevent lying demagogue's from gaining power). Is it "bias" to use a standardized method or algorithm (even one that involves humans reading natural language") to filter content by accuracy and provide some interface indication not just for phishing, and for porn, but also for falsehoods?
- morgante 10y ago
  Realistically, this is exactly how they would go about trying to stop Drumpf. Plausible deniability and all that jazz. In fact, decisions which Facebook makes this election season will impact the result regardless of the intentionality behind those decisions. Even something as mundane as the mix between links and images impacts different candidates disparately.
  - makomk 10y ago
    Realistically, the only way for anyone not to seem like they're going out their way to attack Trump during this election is to completely forget about fact checking, because he lies so repeatedly and shamelessly about things he's already been caught out on: https://www.washingtonpost.com/news/fact-checker/wp/2016/03/22/all-of-donald-trumps-four-pinocchio-ratings-in-one-place/ https://www.washingtonpost.com/news/fact-checker/wp/2016/03/...

kristianc 10y ago

> “It was degrading as a human being,” said another. “We weren’t treated as individuals. We were treated in this robot way.” Strange that people at Facebook would feel apathetic toward individuals that can't code.

drdeadringer 10y ago
Strange that their job was not to code but to do journalistic duties as journalists.
dredmorbius 10y ago
Google has this disease quite badly, from multiple sources. It's pretty common elsewhere in SV as well. And is a cardinal error.

morgante 10y ago

Having worked with journalists, these sound like typical entitlement complaints. Frankly, a lot of writers have an attitude that they're "artists" who shouldn't be rushed and don't have performance requirements. Frankly, it does not sound very hard at all. They have to write 20 posts a day, but each post is only a headline and a brief summary. A focused writer can finish that in 15 minutes. > We had to write in the most passive tense possible. That’s why you’d see headlines that appear in an alien-esque, passive language. Oh no! How dare Facebook strive to be neutral and passive. > After doing a tour in Facebook’s news trenches, almost all of them came to believe that they were there not to work, but to serve as training modules for Facebook’s algorithm Of course they were. Facebook is quite explicitly trying to apply ML throughout their site. Why should they permanently be in the business of employing writers to do something which computers could do reliably and effectively?

SOLAR_FIELDS 10y ago
The article also attempts to paint them as overworked and abused and treated differently because they were contractors. Having worked in the big tech industry as a contractor, their description of contractor treatment and workload is pretty normal (no employee perks, quotas, etc.). This is due more to Vizcaino vs. Microsoft [1] than any sort of abusive contractor agenda. 1. http://scholarship.law.berkeley.edu/cgi/viewcontent.cgi?article=1185&context=btlj http://scholarship.law.berkeley.edu/cgi/viewcontent.cgi?arti...
- morgante 10y ago
  Of course contractors are treated differently. If they were treated identically to employees, the IRS would probably have some questions. They knew they were signing up for contractor position. Frankly, they were lucky to get any benefits at all. The article fails to provide any evidence of abuse though. The worst that Facebook did was not inviting them to company happy hours.
kyle8 10y ago
I shall just leave a link to Douglas Rushkoff here https://www.youtube.com/watch?v=87TSoqnZass https://www.youtube.com/watch?v=87TSoqnZass to explain to you the deep flaws in the "default" techie thinking that your comment is a classic example off.

AndrewKemendo 10y ago

almost all of them came to believe that they were there not to work, but to serve as training modules for Facebook’s algorithm If Facebook is doing it's job and utilizing their biggest talent acquisition of Yann LeCun, then this is exactly what they should be doing. A critical part of the path towards AGI is using humans to teach it.

germinalphrase 10y ago
Thoughts about the utility of schools to serve this function? We are already teaching humans very simple to very advanced tasks and we do so millions of times a day. Is the fact that tasks/knowledge are being structured for paced human learning irrelevant for this purpose?
- aab0 10y ago
  It's not irrelevant, since there is a minor area of ML called 'curriculum learning' which asks how to order examples to teach a ML algorithm most efficiently, and it comes up in some other contexts (for example, a boosting algorithm which focuses on optimizing performance on hard/misclassified cases can be seen as somewhat like curriculum learning, and there are variants of gradient descent which focus on hard examples rather than wasting time on cases where the NN can already get the right answer; and for 'active learning', you want to pick the example which will teach the algorithm the most), but there's not much you can take from known pedagogy at the moment and apply straight to NNs. Even the non-bullshit parts of education like spaced repetition have no clear analogues for tasks like 'train an RNN to write news headlines based an article text using this corpus of human-written headlines'.
- AndrewKemendo 10y ago
  In fact, "expert systems" were built around this principle, though generally more in the industrial context. I don't think that the way knowledge transfer is structured for humans is by default distinct from how we would do it for AI in the long term, but that's a pretty wide range of pedagogy so some tasks make more sense than others for the state of the art today. In fact the way we train ANN/CNN is similar to how infants & toddlers do early classification. Frank Guerin at the university of Aberdeen has done some study in this area that is in it's early stages. As far as I know there have not been efforts to set up schools to train AI's from the general population. The closest thing to it are the efforts by Google and Facebook to train their classifiers (facial recognition, text detection etc...). We are working on an internal process that uses user input to classify objects in the home (furnishings mostly) for auto segmentation and detection.

zxv 10y ago

Reducing the number of journalists implies that supervised learning[0] was effective enough to begin to replace some human effort. [0] https://en.wikipedia.org/wiki/Supervised_learning https://en.wikipedia.org/wiki/Supervised_learning

morning_star 10y ago

why am i reading this this is the most worthless piece of thing i've ever read in my life

dang 10y ago
We've banned this account for repeatedly violating the HN guidelines.
- fallenshell 10y ago
  bless m8

smelendez 10y ago

Maybe Facebook should have advertised this as a temporary "Robot Journalist Training Fellowship"

jccalhoun 10y ago

why would they hire journalists to do this? It doesn't seem like it takes much skill at all to pick a story and write a short summary of it.

jonathankoren 10y ago
I wonder that myself. I think it has more to do with standard corporate over thinking about what's skills are actually required for a job. It certainly make it easier to say, "We want someone to write titles and snippets of news stories. Who does that? Journalists. Cool, let's get some of those." disclaimer: i worked on that fb product.
n72 10y ago
My guess is they needed the best seed they could generate for their ML algos.

n72 10y ago

'Mark Zuckerberg has been transparent about his goal to monopolize digital news distribution. “When news is as fast as everything else on Facebook, people will naturally read a lot more news,”' Oh lord. That's not news. Those are sensationalist, click baity headlines making us all collectively dumber. It's anecdotal, but I'm quite sure the 'news' my girlfriend has gotten from FB has made her less informed. I actually wrote an FF plugin to hide that right bar, since I would now and then get sucked in by that crap and waste 15 minutes reading about Lindsey Lohan or whatever.

morgante 10y ago
Actually, I find that the Facebook headlines are much less clickbaity than the predominant headlines from news organizations.

kriro 10y ago

I don't think the assessment that they were part of a future algorithm is too far off. Given enough expert input picking a fitting image/video or headline should be doable. Worst case you reduce the number of humans needed to one quality assurance person. Algorithm says "this headline, this image"...yes/no;fix.

golergka 10y ago

So, I got interested in this article's description of horrid working conditions and decided to read about it carefully. But I noticed that it gives emotional descriptions of it long before the actual facts. I wouldn't go as far as calling this a manipulation, but it's certainly a disturbing writing style. 3rd paragraph: > grueling work conditions, humiliating treatment, and a secretive, imperious culture in which they were treated as disposable outsiders. 6th paragraph: > “It was degrading as a human being,” said another. “We weren’t treated as individuals. We were treated in this robot way.” And then, finally, on 10th paragraphs, we get a glimpse on the facts: > they received benefits including limited medical insurance, paid time off after 6 months and transit reimbursement (BTW, is it usual for contractors to receive such perks?) > A company happy hour would happen at 8 p.m., and we’d be working Horrible, inhumane treatment indeed. > Over time, the work became increasingly demanding, and Facebook’s trending news team started to look more and more like the worst stereotypes of a digital media content farm. Managers gave curators aggressive quotas for how many summaries and headlines to write, and timed how long it took curators to write a post. The general standard was 20 posts a day. 20 posts during 8 hour work day is almost half an hour on one post. It is considered too little? Seriously? So — apart from all the pretty words, I didn't really see any especially bad treatment. Hell, I'm pretty sure that your average newspaper employees have more nightmare stories.