To be clear, Google isn't accused of intentionally tracking children in defiance of privacy laws, they're accused of not erring hard enough on the side of "this is a child" when guessing the age of a user they don't know the age of. This sounds like a genuine grey area that YouTube was on the wrong side of. Doesn't mean they were right, but everyone's acting like this is flagrant disregard of the law when in fact it is an edge case that didn't yet have a clear standard.
It's even worse. The account signed a EULA saying they are over 13, and this account most likely also happens to be shared between the parent and their kid, so the explicit video being watched aside, it's impossible to tell if the phone is currently in the hand of the parent or the kid, constantly going back and forth.
The solution they're going to implement is to treat anyone watching a "kid" video as a kid. Not only is this going to hit creators hard, but the "kid on their parents phone" could still watch non-kid videos and they would get tracked...
It's a really messy solution. It's like if we started suing porn sites for every minor on their website that lied about being 18.
Google/youtube still won't be able to do anything because kids are smart. They'll just make accounts saying they're 22 years old. It reminds me of the times when I was like 14 and started watching pornography. To say the least, you quickly learn to just click the "yes, I am 18+ years old". That's the reality. If YT/google were to really enforce it they would make all YT/google accounts require ID submission for commenting/uploading etc.
Google already has a minimum age of 13 to create an account (the maximum age before COPPA no longer applies). The issue here is that if a child didn't sign in or used an account where they claimed to be older, they would still have their information collected - the FTC is arguing that Google should have disabled tracking on all videos targeted towards children, even when they didn't know the age of the person watching.
So basically all kid content creators are going to be hit hard once this goes into effect, and we'll get another wave of backlash about "Youtube is killing our channel"
The trickier part is "all age" content. Even something like Minecraft, which at the surface is aimed at kids, I know many adults who watch and play the game... Are all Minecraft streamers screwed?
I don't see anything saying they can't play ads? They can even target them without having to collect any info. If you are watching a kids video its pretty obvious an advert about toys or kids games would be relevant.
That's true, it will still play ads, but obviously non-targeted ads are not as lucrative (an adult watching a Minecraft video doesn't care about toys), therefore, there will still be a non-negligible hit in revenue. If the difference was negligible, why would Google spend so much time developing targeted ads?
The adult watching kids videos is something I did not consider. Target ads are useful for adults because their wants and needs are far more diverse. Only a fraction of adults are interested in buying a circular saw so targeted ads are needed but for kids you can put a pretty safe bet on the fact that a lego advert on a kids show will hit the correct audience pretty spot on.
I don't understand why kid content creators are going to be punished here; Google can't collect data from people watching kids content and can't serve targeted add to them - so what? They can still count views and play untargeted ads and pay out creators?
Nothing about the law or settlement is about blocking children from viewing inappropriate content. It is solely about Google violating children's privacy by collecting information on viewers of content that targets children.
The real problem is much worse than just data collection. The children's content on Youtube is the most vacuous garbage in existence and young children have wasted billions of hours watching it on autoplay.
The content on YouTube absolutely differs from the content on regular television.
Most children's television content has some value: a lesson, moral or at least an actual story.
Others have mentioned the horrifying inappropriate content, but even ignoring that, there is a mind-boggling amount of content on YouTube that is really devoid of value (at least in terms of what I want my kids watching):
* Nursery rhymes -- which is not so bad, but it just keeps going forever
* People (often adults) dressed up as children's story characters, just goofing around
* People (not always kids) literally just playing with toys
* People (not always kids) doing some mind-numbing repetitive task (for 10's of minutes) like opening kinder eggs, molding play doh or pouring glitter on things
Occasionally when staying in a hotel, I flip through the shows available on the cable TV there. With a 3-year-old, that now means flipping through the kids shows.
Unfortunately, these are now unbelievably vapid and pointless, even on PBS. It’s not quite clear what happened. The comparison to early Sesame Street, Mr. Rogers’ Neighborhood, Newton’s Apple, Long Ago and Far Away, Reading Rainbow, Bill Nye the Science Guy, etc. is stark. I’m not convinced most of the current shows are any better than watching an endless stream of poorly produced nursery rhymes on youtube.
I wouldn’t let young children spend nontrivial amounts of time watching “Kids’ TV” in 2019.
Content farms thrive by children's content on YouTube. Something similar could never exist on other curated platforms like TV just as a matter of scale. A content farm can put out 100 ten minute videos of the same type just by swapping out some assets and keywords. At least one of those will get somewhat popular. This is even more devious with MCNs where multiple channels group up but revenue goes to a single entity. If YouTube flags a single channel, it doesn't hurt the entire MCN so you can get around the spam filters that YouTube almost certainly has in place.
Kids shows on traditional tv shows have story lines (conflict, resolution), educational content (numbers, letters, basic math), some level of basic logic/moral dilemmas. And at the very least they didn't have disturbing violent and sexual content that you get from kids youtube.
For instance, I suspect a lot of kids do not have proper fear of wild animals, because from television they know animals are friendly, and some can even talk.
Additionally, bad guys wear black and have scars, while good guys are clean shaven and wear attractive clothing.
Children's television didn't include a nearly endless flood of horrifying and sometimes trauma-inducing autogenerated content that (illegally) leverages pop culture and otherwise kid-friendly characters, all just to push as many ads as possible into kids' faces.
There's always been problematic or otherwise weird children's content, and content that was mostly just an ad (80s cartoons were bad about that), but some of the stuff you can find on YouTube is in a whole new class of questionable. At least some of those 80s toy commercial cartoons were still entertaining in some way.
Exactly, it's unconstrained algo generated content that just seeks out the weirdest stuff that will capture a childs attention. It has been discussed on here before and it is super creepy.
In 10 years people will be looking back at this in shock that it happened and there will be downstream affects on the kids exposed to it. It's literally programming weird stuff into the psyche of these young children.
>In 10 years people will be looking back at this in shock that it happened and there will be downstream affects on the kids exposed to it. It's literally programming weird stuff into the psyche of these young children.
~20 years ago people were deeply concerned about the impact and influence that sophisticated television and radio advertising campaigns were having on people, culture, and the economy.
It's gotten so bad now that we seem to have forgotten about how creeped out people were about the comparatively tame advertising through broadcast media. It's not safe to assume the world will march forward to be better or more moral as time goes on. We're just as likely to regress as progress, and the myth that progress is a passive endeavor makes it more likely that we won't.
Clearly their worries were completely unfounded given how sophisticated, nuanced, literate, economically sound and overall well mannered the current electorate behaves.
People have said the same exact thing about every kind of media ever invented. Books, magazines, television, music, and now YouTube. Do you have anything to actually substantiatiate your claims?
No, you clearly are not aware of what a typical piece of children's content on Youtube is like. I also have a low opinion of the average cable TV kids show, but the two are not even comparable in quality.
It was also exposed that pedophiles were talking to children on youtube through comments and getting them to do inappropriate stuff like telling them to do the "Popsicle challenge" or "stretch challenge" to basically get them to unintentionally expose themselves. These kids actually have youtube accounts, know how to use a phone to record themselves and were basically doing these challenges then immediately after uploading the video to YT. Parents were completely oblivious to this. The parents think their kids are dumber than they really are. "My son? He couldn't figure out how to register a youtube account" etc. meanwhile the kid has been uploading daily for months.
It seems like Youtube Kids is an abomination on every level. Between Elsagate, privacy violations, and the generally pathetic unboxing-oriented nature of its mainstay content, it's a revolting offering for children.
Of all the Google products that get deprecated or killed, this is the one I wish they'd amputate.
I think it's overblown. My kid watches YTK sometimes, and he watches mostly people playing with Lego and hotwheels cars, and animal videos. I've never actually seen any of that sort of questionable content.
"A lot of that stuff they are playing with is sponsored content. The toy maker is paying for the kid to play with that toy."
I think this ship sailed like eighty years ago... Seriously just about every successful kids show from my youth - with the possible partial exception of Sesame Street - were thinly disguised toy advertisements.
> I think this ship sailed like eighty years ago... Seriously just about every successful kids show from my youth - with the possible partial exception of Sesame Street - were thinly disguised toy advertisements.
I grew up on Arthur, Mr Rogers, and Sesame Street. It’s certainly possible to have quality noncommercial children’s programming.
>I grew up on Arthur, Mr Rogers, and Sesame Street. It’s certainly possible to have quality noncommercial children’s programming.
All of those were products of PBS. So it seems like if we want it to be possible it needs to be primarily funded by donors, grants, and public spending.
Perhaps the problem is the quantity? I'm not sure what world we're thinking we should just give our kids a phone and let them watch youtube as long as they want?
I don't think that negates the garbage content and algos from youtube, and we avoid youtube with our kids, but if kids are running out of quality content, send them outside and lock the door.
> I'm not sure what world we're thinking we should just give our kids a phone and let them watch youtube as long as they want?
What we're thinking is "I need an hour free from entertaining my child, this entertains them."
It's quite simple, it's the modern equivalent of "go outside and play" for a generation where kids going outside for hours of loosely supervised play is frowned upon by society.
I'd disagree that it's a modern equivalent for the child. Maybe the same thing for you. There are plenty of times I need a break. It only takes an extra minute to redirect to something that takes effort for the child. Remember when your mom said, 'if you're board, you can clean your room' then you found something to do? Try that instead of handing them a phone.
It's funny that people are worried about letting kids play around their neighbors outside but letting any random person have access to them from YouTube is preferred? I can at least vet my neighbors.
I don't give a fuck what people think. There's no laws against it. I send my kids outside and tell them to come home for dinner. I live in a major metro in the US. They visit friends, neighbor dogs, bring home flowers from neighbor's gardens.
It's hard to believe that you can trust a kid on the internet if you can't trust them outside by themselves. And let's be honest, the chances are actually higher that you're a threat to your kid than some random person trolling the neighborhood.
Where did common sense and doing what's clearly best for your kid go?
edit: if you give your kid one hour per day to take a break, you need to come up with 7 hours of viable content a week. That should be pretty easy to do without going through youtube. There's stuff out there that is reasonable and we haven't even dipped into minecraft.
To add to that list: Power Rangers, GI Joe, Gundam, Care Bears, He-Man, GoBots, Cabbage Patch Kids, Thundercats, TMNT, Street Sharks, hell they even had a show for Rubix Cubes.
Just because a show is built around a product doesn't mean it's a show devoid of value or quality entertainment.
Perhaps you're right. Still, all of the YTK stuff I've seen is strange to see the least. I'm not sure if this is the right word, but a lot of it seems extremely artificial? Even stuff with live actors has that kinda quality to it
This lawsuit is not about Youtube Kids. YTK already has no data collection nor targeted ads (obviously). This is about kids watching Youtube Main on their parents account (you need to be 13+ to have a Youtube account).
Youtube Kids also now has whitelisting, so you can in theory just pick channels manually and avoid all the bad crap.
This whole Elsagate stuff is only the tip of the iceberg. People like Jake Paul produce horrific 'content' targeted at young audiences. I can recommend this video on the topic (not 100% about YT Kids, but it is in there): https://www.youtube.com/watch?v=ywcY8TvES6c
It's described in this video. A child wore a cop costume and did bad ASMR. The YT description was "SASSY Police Officer / Cop Roleplay ️ SASSY AND RUDE" -- uncomfortable language when applied to a young teenager.
If I understand correctly, there was a young teen girl who did an ASMR video in a form-fitting “sexy cop” outfit. The whole thing was YT-monetized, and aided and abetted by her parents.
The even bigger deal was the fact that a YouTuber that made a video pointing out the problematic nature of those ASMR videos had his critical video taken down got a content strike.
It's the only app I found where I can create whitelist of YouTube channels - my kid use it and loves it. And I pay for premium account to avoid all the ads - works great.
Would be great to have an option to "ban" channels in YouTube (without additional apps), because some of "kids" content on YouTube looks pretty insane and I think can be harmful for kids.
I spent a few minutes of DuckDuckGo'ing but couldn't find anything... does anyone know if there's a way to see how/if these fines are actually being paid?
UPDATE: Looks like most fines are paid to the US Treasury. [1][2]
Because penalties for bad behavior shouldn't come out of your tax burden? I dunno if that's what GP meant but it's certainly horrendously unethical enough for me.
Taxes are assessed at a percentage of income. If a fine is deductible, you have lower income, and your taxes are lower, yes. But you paid the government for that right! If your tax rate is 30% and you pay a $1000 fine, you save $300 in taxes. But you paid $1000 for that right!
> If your tax rate is 30% and you pay a $1000 fine, you save $300 in taxes. But you paid $1000 for that right!
What it actually means is that your effective net punishment/fine is $700, irrespective of the headline number. Which matters if the numbers are designed to be impactful.
You’re advocating for having three categories in corporate accounting instead of two: revenue,
costs, and penalties?
Ostensibly the courts would factor in tax benefits to their calculations for punitive damages, under the current system. I don’t believe there are any limits on what punitive damages can account for, beyond “make it hurt”.
Not sure if this is a fine written into law though, which would not be able to easily account for tax differences.
The most common fines and penalties (late fees on federal and state tax returns, are not tax deductible (nor are parking tickets or safety violations).
Insurance is an expense, some insurances premiums are and some are not tax deductible.
Capital Expenses and Equipment, depends, but usually not deductible as a expense but as depreciation over time.
Commuting Costs, generally non-allowable deductions.
Home Office, its tricky and full of bullshit. Same with personal expenses (car/phone)
Political Contributions, tricky and full of bullshit, generally not tax deductible.
Corporate gifts? your deduction is capped at $25.
The Tax Code doesn't have bright-line rules like "every cost" is tax deductible.
That's definitely not true. I'm not an expert in US tax law, but there are a number of important differences between accounting and tax income/expenses:
* Fines, as other people have stated, are usually not deductible. They aren't in Canada, for sure, and they are added back to income (i.e. you're taxed on that amount).
* There is usually a big difference between how leases are treated for accounting purposes and for tax purposes
* Accounting requires that things like accounts receivable to be measured at the expected value. Accounting generally requires an allowance to be booked as an expense for doubtful accounts, even if you don't know which accounts are doubtful. This is usually done on a basis like, historically 2% of accounts are doubtful, so we'll book that expense. For accounting, that's fine, but for tax purposes only actual bad debts are deductible.
* In Canada, capital gains are taxed at 50%, while for accounting purposes 100% of the gain is included in income.
* Capital asset purchases are not deductible, but are depreciated instead over time
* There's limits on things like meals and entertainment. In Canada, only 50% of them are deductible as it is assumed there is always a personal portion to a meal.
So it's not like everything is always deductible, there are many things that income is adjusted for for tax purposes.
not being an accountant as I understand it in my country fines to the tax ministry are not tax deductible. So if this is the case in other countries there would be some legal precedence for saying "fines for this type of offense are not tax deductible"
It depends. If you're a profitable company and the money comes out of a dividend you were going to pay your owners, it's $700 (and the ultimate cost to the owners is less again). But if you're struggling to pay the bills, or the fine is big enough to threaten your operating cashflow, it's a full $1000 you can't spend on new machinery or whatever.
Just a slap on the wrist in the grand scheme of things, considering how large the children's ad market is
>YouTube said that, in a about four months, it will begin treating data "from anyone watching children's content on YouTube as coming from a child, regardless of the age of the user" and will stop serving personalized ads on this content and end comments and notifications on it.
I wonder if spoofing your identity as a child will become a viable anti tracking strategy
Oh boy I'm looking forward to hearing about random YouTubers' content getting flagged as children's content and losing a further chunk of their revenue...
It would work but also you'll face more restrictions or outright blocks. For example, you register an account and put you are 5 years old. After clicking register you get redirected to disney.com or something.
It's an idea that's been around for a long time and one could argue it's growing in popularity as a concept. But it's not very popular in the legal world yet.
Your crimes are against the state in most cases, which is where the fines will go.
> YouTube earned millions of dollars by then using this information to target ads to the children, according to the complaint.
How many millions?
Because if it's $500m then I'm going to pissed off because Google's just going to consider the fine a cost of doing business and continue about their merry way.
But if they only made, like, $2m from this information, then the fine seems like a legitimate tool to stop this sort of shitty behavior.
I tend to think that it'll still be considered a cost of doing business. They'll just figure out another way to do what they want and deal with the consequences if needed.
Their meetings must be like Phillip Morris or whatever it's called now:
"This advertising method work great! What other markets can we apply it to?"
"There is this a market it works even better on, and keeps them coming back for decades."
"Decades? Great! Let's do it."
"Well, that market is kids. Are you sure we should?"
"Hmm... didn't think of that. Well, too late to stop now. Besides, the government set up rules that when you do it, you just pay a fine. It's like a toll road. We'll follow their rules and everything will be fine."
"It's decided then. Project Kids is on."
When some people in Google saw the temptation for such outcomes, they proposed publicly committing to avoid being evil. When push came to shove, they backed down on the public commitment.
With each situation, they can choose to return to their public commitment or what they're doing now.
They seem to pick this option many times and the public commitment keeps getting relegated to more minor places. Last I heard "Don't be evil" was buried in an HR document.
This comment is very disingenuous and misleading. Youtube Kids, the real product targeted at kids, does not collect data or show personalized ads. From my understanding, this lawsuit is about collecting data on kid videos on Youtube Main, which technically you need to be 13 years old to have an account for in the first place.
I guess the argument here was that the parents let their kids watch on their account, and since the video is clearly for kids, Youtube should've not collected data on it because obviously adults wouldn't be watching this video. Seems a bit of a stretch.
How is it a bit of a stretch? If I setup a table outside an elementary school with a colorful sign that says "FREE RAINBOW SPRINKLE BROWNIES" that are unknowningly laced with drugs, and the parents get upset that I did this, I can just say the free brownies were not intended for children?
Youtube knows kids are watching videos. The only thing they don't know is, are the parents present. They could implement a captcha between videos the majority of kids would fail ("What is 25-11?" / "What is the engine of a train called?") but they chose not to do anything like that.
You can assume that some, and maybe most of the people watching certain videos are kids, but there's no way to know for sure since the account itself is for an adult. Just like on a porn site, you need to agree that you are an adult, but realistically a lot of non-adults probably are watching too. The question is, if in the user agreement you agreed that you are above 13, but then you aren't, is that the websites fault?
Now it's up to Google to 1. figure out which videos (uploaded by others) are for kids, and 2. automatically assume everyone watching said videos are kids. Both of those steps are rough approximations. The better solution is to stick to the EULA and have kids actually use Youtube Kids, instead of "guessing" what videos and users are.
> From my understanding, this lawsuit is about collecting data on kid videos on Youtube Main, which technically you need to be 13 years old to have an account for in the first place.
I mean, I would consider a 13 year old to be a kid? And more to the point, you don't need to have an account to have ads targeted at you. YouTube can say "well kids shouldn't be there" but if they were interested in knowing if viewers were kids, I'm sure they could figure it out.
> I guess the argument here was that the parents let their kids watch on their account, and since the video is clearly for kids, Youtube should've not collected data on it because obviously adults wouldn't be watching this video. Seems a bit of a stretch.
You're missing the point of the comment you replied to. Laws that are enforced by fines are legal for the rich, full stop, because they just pay the fine. In the case of individuals, this is why the wealthy are fine with paying speeding tickets and the like: the money means nothing to them. So, you have a company like YouTube making millions and millions (if not billions) of dollars targeting ads at children, probably the most valuable ad demographic because they haven't been browbeaten with advertising messages enough yet and they still work very well on them. So YouTube does this stuff, and sets aside money for when they're caught. Then they pay the fine, and get a tax write-off in the process.
Fines don't do shit. Period. You wanna stop this kind of thing, change the laws so that the management in charge of these kinds of decisions and programs, or failing that, those that failed to see what was being done, are held personally accountable. I guarantee you'll see a hell of a lot less stupid risk-taking by companies when the penalty isn't a pittance of the profits attained.
It doesn't matter what you consider, the law under which they were sued is explicit about the age below which you are not allowed to collect data. This is the reason why almost every website out there asks you if you're older than 13 when you create an account.
> You're missing the point of the comment you replied to.
The comment I replied to opened with a whole mock narrative about engineers deciding to advertise on kids. My point is that this is a non-trivial and non-obvious issue. As far as Youtube is concerned, they made a site where 13yo+ can upload and watch videos. What ended up happening is that people uploaded videos for kids, and others let their kids watch said videos. It's disingenuous to make it sound like Youtube intentionally planned this.
I agree that fines don't do this. And again, the laws are pretty explicit, but this is a complex case, which is what I was trying to convey. It's not an obvious "someone in a dark room wrote code that collects children data".
Narrator: "A new car built by my company leaves somewhere traveling at 60 mph. The rear differential locks up. The car crashes and burns with everyone trapped inside. Now, should we initiate a recall? Take the number of vehicles in the field, A, multiply by the probable rate of failure, B, multiply by the average out-of-court settlement, C. A x B x C = X. If X is less than the cost of a recall, we don't do one."
Businesswoman on plane: "Are there a lot of these kinds of accidents?"
Narrator: "You wouldn't believe."
Businesswoman on plane: "Which car company do you work for?"
Okay? But what does that have to do with literally spreading a lie? I understand saying that Google is evil, but to falsly claim that they've removed "Don't Be Evil" when it's clearly still there makes it just a straight up lie.
> to falsly claim that they've removed "Don't Be Evil" when it's clearly still there makes it just a straight up lie.
I wrote, ". . .relegated to more minor places. Last I heard 'Don't be evil' was buried in an HR document", which doesn't claim they've removed the phrase. It seems consistent with the link above, https://abc.xyz/investor/other/google-code-of-conduct, that shows the phrase at the bottom of pages and pages of bureaucrat-eze (mentioning HR several times), like the ark at the end of Raiders of the Lost Ark.
You seem to imply that I "literally" spread a lie and falsely claimed a "straight up lie." Did I misread?
That’s the problem with vague general principals, they are used to justify occasional acts of tyranny but do nothing to guide day to day decision making. Only concrete actionable guidance matters. It’s easy to think whether a particular action maximizes shareholder value, whereas asking if an action is evil so prone to reinterpretation and conscious or subconscious abuse as to be worse than having no guidance at all.
> I wonder if spoofing your identity as a child will become a viable anti tracking strategy
That's a great idea.
Along that line, I've been wondering lately what are the ramifications for telling dipshit websites I'm European so they must delete my data if I'm not?
And every Canadian learns to press the I'm from the US button on every website (see: car sites) if they want to see the better website and aren't actually shopping.
It's gotten better for Canadians as the average quality of sites has improved, but it was consistently designs from a prior era and 1/10th the content.
Companies that operate in multiple regions sometimes use these buttons to make language-specific or region-specific content available. For example, some cars may only be available in some regions or may have different branding. Another example is how the Apple Store has region-specific editions. Also Netflix filters the library by region.
A lot of eCommerce websites will only ship to/operate in the U.S. for a variety of reasons (business registration requirements, consumer protection laws, currency exchange, international shipping, etc. Hence websites like newegg.com and newegg.ca.
Certain product lines or pages will be missing, or the details will be less informative on the Canadian/International regionalized website. Not to mention most online shops will only ship to the US.
Seems a bit backward to me. Why should we care so much more about kid's privacy than adult privacy? Are you really more concerned that people know about your Barney the Magical Dinosaur addiction at 8 than your sordid affair at 32? Would you much rather keep secret that you walked back and forth from kindergarten everyday than your employers confidential design files?
Adults and teens have secrets, kids relatively speaking for the most part don't, at least not ones worth protecting through strenuous tech laws. Okay okay, I get that we don't want them vulnerable to be manipulated into stealing their parents money. And Google is as always scrum. But the general laser focus on kids when we should be concentrating on everybody seems just like another think of the children hysteria.
This account has been breaking the HN guidelines by posting unsubstantive comments. Also, the username is borderline trollish, which isn't allowed here: https://hn.algolia.com/?sort=byDate&dateRange=all&type=comme.... I've banned this account, but if you don't want it to be banned you're welcome to email us at hn@ycombinator.com.
I don't see how anyone could come to any other conclusion. It feels like every day we find out about yet another tech company finding a new way to extract more of our personal data. At this point, the only appropriate response in my opinion is to ask: are we mad enough to do anything about this yet?
170M fine for a company that makes ~10B of earnings and 30B of revenue per quarter. That's like someone making 100k per year and only having to pay 50 cents in fine. Even parking fines are a much more steeper in % for average joe.
A fine of anything less than a billion is like breeze hitting Google's face. You want to slap them, make that stock hurt.
