What I’m really unclear about is whether providers are required to use an actual ID to age verify. Does anyone know?

The bill summary on the Florida senate webpage says:

> Such commercial entities must verify, using either an anonymous or standard age verification method, that the age of a person attempting to access the material harmful to minors satisfies the bill’s age requirements.

It sounds to me like “anonymous age verification method” could just mean that the website asks how old you are? What constitutes verification here? That sentence makes it sound like they can choose to use whatever feeble method they want.

At face value this law seems like political points being scored by passing a widely popular law that changes very little in practice (bumping the minimum age from 13 to 14).

Anonymous verification could be something like OAuth. Government run or certified probably. You'd need to provide an ID to OAuth provider once, but the actual service requesting verification would get as little as your age and email.

I just wonder if they even have to go that far. I didn’t really see much of a standard of what is age verification defined.

> Anonymous verification could be something like OAuth. You'd need to provide an ID to OAuth provider once

That doesn't sound very anonymous to me

How many social media sites allow you to do anything without an account? Twitter used to be wide open but X competely locked down. Instagram lets you click 2 things and then the paywall pops up. I'm not sure about Facebook but it isn't much better.

Without speaking to merits, the entire point of this law is to reduce usage by minors. So being able to do less without an account is a feature through this lens.

The question was about merits though, you can't just ignore the question.

How do teenagers find and discover businesses when they are locked behind a sign up wall for social media?

> How do teenagers find and discover businesses when they are locked behind a sign up wall for social media?

Why is this even a problem?

Oh no, businesses can’t target advertising directly at children?

Who said anything about advertising?

This is about businesses using Facebook as their main website. Where do ads come in?

Are teenagers not real people or something? They don't buy things or shop?

Kids can “find businesses” the way they always have

Like, with ads? That you were so against above. What's wrong with them going to a businesses online presence?

This is an unfair parallel, but it's like worrying about businesses who advertise on pornographic sites or on cigarette boxes. Allowing visibility to businesses who advertise on a certain platform could be judged (by society, law, and voters) to be outweighed by the need to restrict youth access to social media. There is a variety of precedent in society and law in restricting youth access to something that is otherwise deemed legal. Just because businesses advertise on social media (in my opinion), the decision to restrict access should not be altered by that advertising strategy.

They don't advertise on FB, they host their main web presence on FB.

That's like saying you go to Pornhub to buy Manscaped Hair Trimmers. You don't. They _advertise_ on PH, but that's not where you buy them at.

Why is everyone replying to me about ads? Who said anything about ads?

It will be interesting for businesses like restaurants that don't have paper menus but have you scan a code that opens an IG with their menu in an album. Frankly, I applaud the state for eliminating this use case.

Twitter required a login long before the rename or Elon owned it. He actually removed that restriction for a while before reenabling it.

And with some simple div removal, IG doesnt require a login to view content. This is true about a lot of the paywalled sites.

That doesn't sound true to me... Idk specific policies, but my experience was that I was never logged in to twitter on my desktop and I was never login-walled out until recently under elon.

edit: https://www.cnn.com/2023/06/30/tech/twitter-public-access-re...

> People without a Twitter account or who weren’t logged in used to be able to scroll the platform’s homepage and view public accounts and tweets. But as of this week, when such a user opens the platform they are met with a screen prompting them to sign up or sign in to Twitter.

Ive never had a twitter account. And I was definitely not able to view most content during the pandemic.

from 2021: https://www.reddit.com/r/DataHoarder/comments/pa6dra/twitter... and from 2022: https://news.ycombinator.com/item?id=30615371

A plus of this law would be if this would force social media sites to stop locking down read access behind a sign up wall.

I think the barrier to entry with creating a new account on each site right now is low when no ID is required - so social media sites lock everything down.

With the new friction of requiring ID, it could be harder to get users to create accounts so locking down everything won't make sense from a viewership access perspective.

I don’t have a Facebook account, but browse business Facebook pages without trouble pretty often.

The last thing we need is to force everyone to have ID to “protect the children” before I can go on to a site on the internet.

Which is not what anyone is calling for.

Pornhub just blocked Texas. We are very much using children as an excuse for big government.

Big government forced licensure upon motor vehicle operators because early on they demonstrated a tendency to cause societal harm. There is a place where libertarian ideals have to be curtailed for the greater good.

You didn’t answer the question. It’s one thing for the government to have your ID and tied to a physical activity. Do you want Facebook to have a copy of your ID? Do you want the government to be able to ask Facebook who said something that offended them for your real ID?

There have been cases where the cops arrested someone for criticizing them on Facebook

https://theconversation.com/mocking-the-police-got-an-ohio-m...

The cops also raided a newspaper for criticizing them

https://kansasreflector.com/2023/08/12/in-marion-county-news...

We have to be free to criticize the government anonymously

The law appears to require Facebook offer "anonymous age verification" through a third party (so ID is not shared with Facebook), and requires that third party must not retain or share PII used for verification.

For reference: https://flsenate.gov/session/bill/2024/3/billtext/er/pdf

The Texas porn law similarly required no records be retained, with pretty steep penalties ($10k per record) for noncompliance.

What are the positive effects of blocking Pornhub?

What? That's literally what we are talking about!

No, we're talking about stopping children having social media accounts.

How do you imagine that will be enforced? Perhaps by... ID?

Do you have another method to prove age? One that doesn't require ID, and can be implemented as of today?

> How do you imagine that will be enforced? Perhaps by... ID?

Emphasis mine. It's that that which I'm disputing.

This law does not require you to have an account to go onto a website. It is a law that requires you are age-checked to hold an account on social media platforms. These are different things.

Holding an account —broadly speaking— allows you to post, like, comment, follow and be profiled in an enduring fashion. Some of these can have life-long implications that 12 year olds aren't equipped to evaluate. How many adults can identify unhealthy social media usage and do something to stop it?

I don't know how age verification will work in practice. Checking a government ID is one way but if you're handing that off to a third party service, it's expensive. They don't need to know who you are for certain. A similar check is when you buy alcohol. If you look young, you're asked for ID.

When I buy alcohol, I'm not because I look like a man with kids nearing his forties. It would be far cheaper for Facebook to have an AI watch a video of you speaking to camera, with ID as a fallback. Many social media platforms are ingesting gigabytes of our faces every day so. They probably already know the rough ages of their users.

But going on a website (eg Youtube) doesn't require you have a profile. Going on Facebook business pages doesn't require it. If anything, laws like this might mean that things like Instagram have to be more open to preserve their reach. Not a bad outcome, IMO.

When you buy alcohol, it’s very easy to see whether they are taking and storing your picture. You can also buy alcohol with cash leaving no digital footprint.

And is AI the new blockchain that will magically solve every problem? And most children over 14 don’t have government IDs either.

If you don’t want your kids to be on Facebook, use parental controls. It’s the parents responsibility.

I live in Florida for context. The same government officials talk out of the left side of their mouths about “parental choice” and now this law says children can’t be on social media even with parental consent.

Parental choice exists where parents can make informed choices.

You don't know how much their child is being affected by platforms that have been designed to be addictive, that go out of their way to spur conflict and argument (sorry, "engagement"), and that have so few parental controls that it's near impossible for a parent to actually vet what's going on, who's interacting with them.

You're not even allowed to give your kids alcohol. There's clearly some precedence for protecting kids from their parents' inability to make informed choices.

And ML models for detecting age have existed for a long time. It's not magic unicorn dust. It's just one way to avoid this ID stuff that everybody is upset about.

Then educate the parents and tell them about parental controls.

But would you be okay if I used that same argument about why open computing platforms are bad because adults don’t know enough about computers to avoid malware and ransomware?

And how do you know they're not a child?

Bars do it by having you show an id.

Online pot dispensaries do it by having you upload an id.

Texas expects porn sites to do it by having you upload an id.

How does Florida expect a site to do it?

This is a legitimate question that I want the answer to. Presumably "check this box" isn't going to cut it. So if it's not the most common way to enforce an actual legal restriction, then what is it?

There's really no need for ID checking. Most porn sites already self-regulated by marking their content adult with meta tags/headers.

Parental control software picks up on that. [1]

Social media could do the same thing: make a social media adults-only meta tag for parental control software to use.

For the parents that care, and use parental control software, the ID laws won't stop their kids running into porn. The porn their kids are going to encounter is going to be on non-porn sites like twitter or reddit (or small sites that don't care about these laws anyway).

Maybe we needed a bigger push for more awareness or better parental control software but the ID law push is weird and unamerican to me.

[1] https://www.rtalabel.org/

> Parental control software picks up on that. [1]

That's client side. This law specifically makes this a server-side issue; the service cannot let a minor make an account.

How do you do that without ID?

By not having this law and doing it client side. My opinion is the law is bad.

This law might have never come up if we had better parental controls at every level of the software and hardware stack.

I'm a parent I have parent controls setup on my child's devices but it's very hard to dial it in properly and cover all the bases. It should be far more straight forward to manage than it is now.

I'm obviously pretty tech savvy and I would say 99% of parents are not going to get this right.

I can’t speak for PCs or Android devices. But iOS devices already have parental controls built in as do Macs

iOS parental controls are awful. The only way to get any decent control for my concerns that the moment is using the downtime features all day long. Effectively I've told iOS that he should be in bed for 23 hours and 55 minutes a day. It's ridiculous and extremely limiting.

Microsoft's parental controls are sort of ok. I also have separate control software for Windows and I have controls at the router. Of course, none of things can talk to each other to create a coordinated plan (say giving X number of screen time hours per day).

iOS controls effectively block adult sites and you can manually add sites to block and not allow apps being installed without your permission. What else do you need?

Some control over when certain apps can be used.

That has nothing to do with “protecting the children” from sites you don’t want them to see or apps you don’t want them to use at all.

But does have everything to do with parental controls.

Besides the easiest way to know what you kid is doing online is to watch them -- much easier to do if I control when they use it.

A solution to these issues is for the child to not have the device in the first place. A desktop computer in a central place with eye on it can go a long way in managing online activities.

As a parent, as a former child, as someone that grew up with computers in places just like you describe, let me say, in my personal and professional experience… lol

You’re right about parental controls, especially apple ScreenTime. I’ve used computers almost my entire life, I even work for Apple, and I still resorted to calling tech support about it. As far as I can tell, they don’t actually do anything useful, and instead just get in the way.

I do disagree that a technical solution could have avoided laws like Texas. It’s not about “protecting the children”. It’s never about the children. It’s just censorship. It’s just easier to go after a porn site than it is a library.

My sibling in the universe, we are talking about an actual LAW. WRT Texas, a law that is on the books and enforced TODAY.

Get informed.

I've just answered in a sibling thread.

I'm not asked for ID when I order a drink. The bartender takes a look at my ugly mug and makes the call: I'm [painfully] clearly over 25.

Facebook has more than enough processing power to have an AI watch you reading a script straight to camera for 30 seconds to work out a rough age. If you're within 5 years of their idea of 18, surrender that ID, the same way you would if you were in a bar. Don't want to? Don't maintain a social media account. Don't have that drink.

The alternative is setting up a government-maintained 0Auth-style hand-off. They know who you are. The social media site could open a verification ticket, you authenticate with your government and they sign your ticket without the social media site getting any of your details. The trade-off with that is cost and your government now knows you're on TikTok. For some people that last one matters.

If you paired these laws with strong PII protection (see GDPR) to stop social media sites storing this stuff indefinitely, using your data against you, it might be an easier sell.

Thank you for answering.

You have a bit more faith in a technical solution working at scale than I do, but I have to point out, that after repeatedly claiming that no one wanted an ID scan, your proposal involves an ID scan.

I also have to point out, that PII and GDPR protections are meaningless here. Under a government mandated censorship regime, the threat is the government, not some data broker somewhere.

The reason I have faith in a pure-technology solution is because setting up ID-linked database APIs will take time and querying it will cost social networks real money. When this was discussed in the UK, we were looking at 10p per check. That's $1m for Florida alone plus engineering to get it in.

They have innumerable GPU cycles they could redirect to a ML solution, and more importantly this is something that ML and facial recognition systems have been doing in prosumer-grade hardware for well over a decade. If I could set something up to do this in an afternoon, I believe Facebook probably could too.

My proposal only involves ID if you look young enough that the AI isn't certain or you don't want to upload a video. It's cutting out ID checks for 99% over 25yo.

But I do see this more about child safety than a censorship regime. Social networks are poison, even to adults. They're designed to trap us there, keep up "engaging" and not wander off to another platform. Our ability to make an informed choice about whether our kids are being damaged by them (and the people on them - who we have no control over) is seriously diminished. I've seen the stuff my daughters' 8yo friend uploads and the comments would change your mind. Yes, I'm biased, but I don't think that makes me (or this) wrong.

“Think of the children!!!” Has been used for a very very long time to enact terrible laws and quietly remove rights. It is everywhere you look.

Right, definitely wouldn't want that. But do we just have to accept the negative impact the internet can have on children as a necessary evil then?

So you would be okay giving your ID to Facebook or any other site before you can access it?

No, I wouldn't. I'm just wondering if we have to accept the consequences for children or if there's some alternative solution.

How do you propose that social media sites verify your age without showing them your ID?

By showing a third party their ID that doesn't share the actual ID with eg, Facebook. That third party would then share their status with Facebook. Facebook doesn't get the ID, but does get to verify that they're of age (or not).

So now the government still has a way to match a user with a real person and we have to trust a third party company with our ID?

And I should have to do that as a 50 year old guy with grown kids for “the children”?

How about if parents are concerned about their kids, they should use the parental controls that are already available.

And this isn’t theoretical for me. I live in Florida.

If we have to accept the negative impact of bullets in our childrens' bodies from spree killers and cops as a necessary evil to preserve the Second Amendment then yes, we should accept the negative impact of the internet as a necessary evil to preserve the First.

Kids-only internet (moderated by child development phds, idk) you need specific, cheap hardware to access. Less walled garden, more sandbox. It wouldn't be a place for entrepreneurs.

Registration and access maintained at the county level or smaller so that community standards and relationships shape adoption and use.

Low age cutoff with actual adults trying to connect put in jail and on a list.

I’m struggling to parse anything you said.

I'm arguing for a completely separate kindernet. While on packed public transportation.

Who is going to pay for it and who is going to decide what’s appropriate? If the government decides, what happens when a party is in control of government that has opposite views than yours?

> You'd be able to view a business's information, watch videos, etc etc etc, just not create your own.

so...everyone - that Facebook or whatever considers to be in Florida - has to provide ID to post, then?

This is not true! I was looking at a hair salon and bakery recently. Both, being run by millennials, have nothing but a google maps listing and an Instagram account. But I don't have one and after looking at a few photos of cakes and hairstyles, it gives me the boot and asks me to sign in to see more! Adding mandatory government ID to that is crazy.

They sound like incompetent business proprietors. When did having a website become something exclusively for old people? Millennials are in their 30s and 40s.

> When did having a website become something exclusively for old people?

It's more that having a website never became a thing for many businesses in the first place. 30 years ago they might have had a listing in the phone book. Nowadays that kind of business might have a facebook page, or more recently an instagram account. Creating a website (even with something like wix) requires a level of skill and effort above that.

They're just being cheap. Facebook is free (and "everybody has Facebook") whereas Wix and Squarespace are $12-$15/mo for businesses.

For a lot of businesses, a website receives way fewer views than any one of their top-3 social media sites.

I’ve seen a lot who are still paying for a site but rarely bother to update it. All the traffic’s on Facebook.

It is not just 15$ a month. You need to pay someone to develop the website in the first place and then deal with all kinds of issues that eventually arise (updating the underlying framework and/or getting hacked and dealing with the consequences, solving random stuff like expired SSL certificate, etc.).

Idk how something like a cafe would necessarily benefit the business to have a dedicated website. A facebook (or google maps) page presents relevant info such as location/hours/menu in a predefined schema, so neither the owner or the customer need to worry about formatting it (or understanding the format).

Their competence is measured by their success, not your opinions. And it turns out that in many professions, an insta is the thing you need for success.

If you don't like it, that's your problem, not theirs. You're part of a small enough group they don't care about. I'm not sure how that's an age question in the first place, though. The "being run by millenials" throwaway by GP is just... well, at best, sloppy thinking.

There are GenZ businesses websiting, there are boomers insta-ing. You pick the tool for your niche.

Well, I went to the bakery and I can attest that they are indeed run by millennials. I will argue with the technical ineptness point made earlier though. If you can figure out how to make videos with music and edit in animated graphics using the tools Instagram or whoever else provides, you can figure out Wix's website builder.

Why do you need them to have a website? A thing that has no traffic, no discoverabilty, needs to be updated, paid for, etc?

All the information is already on google maps and IG/Facebook.

They have 2-3 billion people in their target audience.

Well presumably a website would be accessible to everyone, account login and government ID or not.

I guess that’s my point though. isn't it a very rational BUSINESS decision to go some where 80% of people and probably 99% of their target market are on these platforms that provide them with free distribution? Outside of the moral qualms of that 1% of people that don’t have a social media account or use google?

Sure. Would you also remark on their hair color? Shoe size? First letter of their name? Because those are likely about as relevant to their choice of advertising medium. It might be true, but there's no relevance.

And the point isn't if Wix is hard or not. It's that they answered the question "where do we need to advertise" for themselves, and just because you don't like it doesn't mean it's wrong.

Placing your online marketing behind an auth wall (Insta) is the definition of incompetence.

Anything that contributes to funnel friction for those wreckers is a social net positive in my book.

Instagram gives you like half a scroll.

Facebook is more generous.

That's kind of on Instagram for forcing you to log in—I certainly never use the site to look at anything. Google maps certainly doesn't require this.

Regardless, I won't bemoan the demise of either business. We need something simpler to drive traffic.

> I'm in no rush to show Zuck my passport

I agree, but I think your problem isn't Zuck, it's with passports.

Government issued licences aren't fit for purpose any more. They were when all you did with your paper drivers licence is show it to the police, but now they've become a form of ID you show man+dog who gets to scrape a whole pile of into from it that can be used to track you. For example, they can follow your passport number or drivers licence to connect a series of what should look like unrelated transactions.

As an example, now when a car rental company wants to know you have a valid licence they demand a copy of it. If you have an accident they use the copy to prove they verified you are licenced to drive, if you do something illegal they can hand over your ID so the police can chase you down. FIDO / WebAuthn / PassKey shows how those things can be achieved without leaking all the information on the licence. It can hand over a one time token saying you have a valid licence and signed by something that chains back to a public key held by the government. The token reveals nothing more than that to the car hire firm, but should they hand it over to the police they can decrypt it to identify you.

These tokens are useless if stolen. They can't link you to other transactions and don't identify you in any way, and yet are far more secure than a bunch of unsigned pixels. In other words unlike a copy of a passport, mostly harmless.

Hard agree. I think making significant progress on this problem would be time consuming - but ultimately a significant win for society.

Hi Seph, any updates on the paper on Replayable Event Graph, looking forward to reading that paper.

Its happening. My coauthor wanted to make some updates before sending it out to people. I'll send you a draft soon, I promise! Its just coming together slower than I'd like. Always slower :S

> don't identify you in any way

That's a critical flaw to those in power.

> gets to scrape a whole pile of into from it that can be used to track you

I can't help but feel your argument is with data protection [and the broad lack of it in the US] rather than government-held databases.

I have a couple of online government authentication methods. There could easily be an AVS API where a website kicks me off to to my government, and they sign a request for age verification, all with very little cost and fuss. That obviously causes uproar from people who think my government doesn't suspect I touch myself when they're not looking.

And a suggestion I've made in a couple of other sibling threads is having an AI watch you reading a script in realtime. I had to do this for a mortgage application a few years ago. Probably cheaper than a government API, and a high success rate on 25yo+ similar to facial-only checking in bars.

> I can't help but feel your argument is with data protection [and the broad lack of it in the US] rather than government-held databases.

It goes deeper than that. Recently in Australia we've had two data breaches, one from a Telco [0] and another from a credit card company [1]. Both were required to collect ID by law, so they gave you a portal to upload photos of drivers licences, government medical insurance cards, credit card and so on. In both cases they leaked the lot.

To say they were unpopular was an understatement. Perhaps 10% of Australia had to get new drivers licences. The were hauled up to front senate committees, CEO's fell on the sword. A lot of political theatre in other words, but while this "take a copy of a government licence as a form of ID" madness continues it will keep happening.

In other news, social hacks against the government electronic ID for their website were used to collect around 1/2 a billion in fraudulent payouts (tax refunds and the like) [2]. And a few years ago the tax office was done for $2B or so for ID fraud waged against our VAT collection. [3] That one was perpetrated by thousands; the instructions went viral on TikTok.

We live in a digital world now, where it's easy to take a copy of any bag of bits. Relying on ID's that don't mutually authenticate and vigorously protect the information they do hand out is downright dangerous.

[0] https://en.wikipedia.org/wiki/2022_Optus_data_breach

[1] https://asic.gov.au/about-asic/news-centre/news-items/guidan...

[2] https://theconversation.com/the-500-million-ato-fraud-highli...

[3] https://www.theguardian.com/australia-news/2024/feb/13/ato-s...

But current trend is websites closing up for browsing without an account, because of AI.