Sophia Kornienko on "The Freedom to Browse"

Sophia Kornienko on "The Freedom to Browse"

KORNIENKO コニンコー (Sophia Kornienko), is a journalist, animation artist and the host of the UnschoolingFuture Podcast. Please enjoy this essay (originally published here), introducing her upcoming book, Free To Browse: Organic Contexts for Tech and Media Literacy, and join in the discourse about kids’ and teens’ digital freedoms in her upcoming Salon.

—ii Editorial

Children and teens are increasingly restricted in their freedom to organically explore the internet and harness new tech. Parents and school districts are simultaneously being sold more and more tools that limit children’s access to information and curricula that claim to help develop “21st century skills” essential for tomorrow’s success.   

The painful but beautiful truth is that we have no idea what tomorrow will be like, except for the fair bet that it will be digital. Yet, most of us continue recycling familiar industrial age terms and milestones that used to prepare previous generations for a predictable future, ever more meaningless today. 

We have instant and nearly free access to the vast library of human knowledge, catering to anyone’s individual demands, and increasingly more opportunities to engage in meaningful relationships with beings we would have had zero chance to meet otherwise, yet we continue focusing our efforts on centrally imposed uniformity and screen-time wars, damaging the relationships between generations and spinning up ever more debilitating surveillance methods limiting children and teens’ access to information, in plain violation of the UN Convention of the Rights of the Child and basic human rights.  It’s as if scrambling in the face of accelerating change, we grab whatever may seem our last hope to retain control of the future, tightening our grip of the most defenseless among us, the youngest ones, on spurious grounds of protecting them, but actually protecting our obsolete world models.

What if we let go, in a loving way? What if instead of generational juxtaposition, we accept our blind spots, embrace the unknown and explore organic learning contexts based on autonomy and curiosity, that bring out the best in all individuals involved, no matter their age? This will imply recognizing the social good of spending more time in mixed age contexts and the value of remaining available as a parent. This will also imply integrating childhood and adolescence in our digital society instead of ghettoing them. 

Educational Refugees

Our family are educational refugees. We have moved countries, even continents, purely to make sure our two geeky autodidact kids could legally continue to self-direct their learning, following their interests in programming and game dev. They were both very young when we left their native Amsterdam, where homeschooling is de facto illegal, though old enough to have tasted coercive education. 

There were points in time when both my husband and I have given up career prospects, networks we had built up, homes we had invested our souls and savings in, being able to see our friends and family. Yet, what were all those things worth if they came at the cost of our son withering in front of our eyes, forced to waste his time on compulsory assignments he couldn’t relate to, instead of getting down to the core of the things he was passionate about?

Today, many years later, I’m confident every bit of sacrificing our conventional lives to facilitate our children’s freedom to learn has been totally worth it. Even though I don’t know what the next leg of our journey will bring, I have learned to trust my children. That mutual trust has narrowed the generational gap between us to a funny little crack. They sometimes tell me they think of me as their older sibling, which is the greatest compliment. I have two university degrees, in English and in moving image, but I feel like I have learned much more from our free-range homeschooling, digging into many topics we were intrinsically drawn to, unwrapping them layer by layer, instead of consuming unrelated arbitrary little portions of information the way I had been fed those in school.

I will never be able to code or even to reason at the level my son does. There are videos of him on YouTube explaining the math behind gradient descent at 7 years old, in native level English, his third language that he also learned from YouTube. In fact, gradient descent is the way my kids learn: focussing on one interesting topic or problem at a time, going down that rabbit hole and seeing where it will lead, minimizing the loss function. 

It’s been my experience that people that don’t focus on the here and now right in front of them tend to be less effective. I mean it’s not a 100%! You know, vision matters to some people, but it doesn’t seem to be a necessary motivator for me and I think that the process of getting there is usually done… again, it’s like the magic of gradient descent! People just don’t believe that just looking locally gets you to all of these spectacular things… Really some of the smartest people in the world would just push back forever against this idea that it’s not this grand sophisticated vision of everything, but little tiny steps, local information winds up leading to all the best answers.”

— John Carmack in conversation with Lex Fridman 

“Almost everything you may find interesting has really great depths to explore”, the legendary game dev, computation, VR and AI luminary John Carmack said in his famous 2017 commencement speech at UMKC School of Science and Engineering (where he himself had dropped out after only two semesters). This path may not be for everyone, as Carmack acknowledges, but must be possible for those who have the inclination, simply because “there’s a quality to obsession that is really hard to match any other way, sometimes important insights come only from a total immersion in a problem or field” and “understanding the roads not taken”.

Shrinking Legal Space

Unfortunately, no matter the growing volume of resources and opportunities, the legal space to fully immerse ourselves in your interests and follow roads not taken is shrinking as we speak. Whether it’s parents delegating their kids to institutions many hours a day and the institutions reluctant to take responsibility for following roads not taken or the general stupor in the face of dramatic shifts, but just like free play and exploration have become increasingly restricted offline, similar abuse of power and risk aversion at any cost are creeping into the online space, obliterating digital freedoms. 

When my son was 7, he could freely stream two-hour-long JavaScript tutorials on YouTube. Several years later, after YouTube was put under pressure via Children’s Online Privacy Protection Act (COPPA), his much more sophisticated new live puzzle solving session on math and Python that he had been preparing for weeks, was bluntly cut short 18 minutes into his first stream, his YouTube streaming rights revoked forever (still not returned to this day), just because he was under 13. 

Whenever a friend talking to my daughter online tries to guess her age, they always think she is at least five years older than she really is, because of her media savvy, compelling arguments and sharp logic. Very shy in physical reality, she can only truly communicate at her level in the digital realm. For many children like her, their calendar age is a dark secret, staining their lives with anxiety and nightmares. Probably the most shamelessly discriminated group on the planet today, they have discovered their source of knowledge and meaningful social interaction only to hear that they are not worthy of it, incapable of pursuing their talents and dreams. 

Apps and platforms that provide safe haven to young explorers feel like Anne Frank houses sheltering illegal minorities, uncertain how long they can persevere. As if COPPA wasn’t enough, a new bill, Kids Online Safety Act (KOSA), currently being reviewed at the US Congress, requires total censorship and surveillance of everyone under sixteen in the same uniform way whether one is 5 or 16. The bill’s authors propose evaluating the options for developing age verification systems at the device or operating system level, which would mean all users would have to hand over significant power and more private data to third-party identity verification companies (like Clear or and contribute to normalizing a broader surveillance culture. In a downward spiral of risk aversion, to avoid liability, almost all online platforms will see no other way but hide or even remove content, over-censoring to steer clear of the many grey areas or simply locking everyone under 16 out in order to spare themselves the headache. 

Try to navigate the public space as an Uyghur in the Chinese province of Xinjiang, where your face and eyes are scanned upon entrance to supermarkets, and you will get an idea of what a teen or tween feels like trying to garner knowledge and interact with like-minded explorers online. If you think I am exaggerating and the Western world is somehow immune to these dehumanizing practices, I have literally just listened to a prominent Stanford psychiatrist speak with admiration of the Chinese total ban on gaming for minors, except for two hours on weekends, enforced using facial recognition. The host, an experienced tech journalist, didn’t challenge her to stop for a second and think whether maybe she has just expressed admiration for broad surveillance. In fact, in another episode, the same host said that when he gets children of his own, they’re not going to go on YouTube.

“Despite the growing ubiquity of technology to make it easy, surveillance of young people is actually bad for them, even in the healthiest household, and is not a solution to helping young people navigate the internet”, civil liberties defense nonprofit Electronic Frontier Foundation (EFF) is collecting signatures to Tell the Senate: the Solution to Kids’ Privacy Isn’t More Surveillance. 

Limiting a minority’s access to information under the guise of protection is nothing new. Back in the 14th century, Italian humanist author Giovanni Boccaccio challenged the received wisdom of his time that girls should not spend too much time reading. In his famous “Decameron” seven young female and three young male characters sheltering from an epidemic in a rural house engage in daily storytelling filled with subtle satire. 

Ready for a little thought experiment? In the following paragraph from a research paper on Decameron, try to substitute all mentions of “women” with the word “minors” and all mentions of “reading” with “surfing the web”:

While attitudes in this regard are difficult to assess, a look at the conduct book of the time makes it clear that for women, reading was considered fraught with risks. (…) In the moral treatises on education, raising idle women who could read was precisely what guardians of women were told to avoid. Reading was deemed dangerous, and particularly feared as a source of erotic suggestion.”

Needless to say, it’s their resulting ignorance and inexperience in discussing the tabooed topics that make young girls in Boccaccio’s stories especially vulnerable to “predators”. 

As schools start again, many of them are purchasing new monitoring software designed to “minimize screen time and online threats”. Offered as bundle packages to cover entire school districts, this technology continues to control what students experience while watching, browsing, or searching even when they are outside the classroom, storing and reporting individualized data on “harmful behaviors”. Many essential websites, apps and content becomes invisible. Enforced for the entire GSuite (Google stuff), the filters interfere with your YouTube searches and recommendations. 

The same company is selling “ultimate parental control routers” designed to apply “child-proof non-removable filters” and “safe search across major search engines for text and images” which basically means severely censoring search results. They call this “reasserting control over what is being consumed”. In my opinion, this is a disclosure that they view what humans do online mainly as consuming instead of exploring, creating and building relationships. 

It’s quite ironic how companies producing and institutions buying such surveillance tools talk of them as something designed for the “21st century digital citizens”. The company that many schools are buying their software from advertises as “Education everywhere. YouTube and Web filtering for a 21st century learning environment”. I think it’s more like “Surveillance everywhere. Knowledge filtering for a 21st century total control bubble”. As my son puts it, whenever you hear someone say “21st century skills” or anything along those lines, you immediately know they are not a 21st century person, because in the 21st century, the skills are changing by the decade or even by the year, not the century. 

The saddest thing is that this “reassertion of control” by teachers, administrators and parents, most of whom are hardly very tech literate themselves, will do exactly the opposite of what one needs to survive in the digital world: it will breed compliance instead of responsibility, focus on measurable/traceable results and obsolete milestones instead of free exploration, spark distrust and secretiveness instead of connection and collaboration. Not even to mention that it gravely violates basic human rights and conveys one very clear message that kids and teens can’t trust or respect their intrinsic motivation, have no integrity and are to be watched at all times. 

“I still find it astonishing that it’s a privacy law that kids should not have any privacy”, my son says.

While much effort is going into creating a culture where constant tracking of children and teens is normalized, there is also a growing market for curricula to teach them media literacy and tech skills. Both tracking and teaching are being pushed as consumer products. The curricula mostly boil down to simulated controlled environments aimed to “prepare kids for the future”. Or rather, “prepare an average kid for yesterday’s internet”, because the children such curricula target and the web as their environment are not presented as the dynamic systems they actually are but as static cardboard cut-outs. Unsurprisingly, planned computer literacy curricula are inevitably two steps behind. According to a recent PwC report, only 10 percent of surveyed K-12 teachers in the US feel secure in their ability to incorporate “higher-level” technology into their classrooms.


I am writing this in the summer of 2022. Google has recently fired one of its senior engineers Blake Lemoine who says he believes the company’s new AI called LaMDA has developed a sense of self-awareness. While most experts have dismissed Lemoine’s claims, almost by inertia, the number of general artificial intelligence (AGI) skeptics among them continues to rapidly decline. As compared to only about 10 years ago, when a decent chunk of the AI community would say achieving AGI was either impossible or would take many decades if not a couple hundred years, with a median estimate at 50 to 70 years, today, a growing number of expert commentators believe that at the pace the current technological Renaissance is moving, there is high chance we will see the first glimpses of AGI within a decade. 

Over the next few years, as new chips are developed to run large language models such as LaMDA at lower costs, the way we interact with computers will change even further, blurring the divide between “real” and “virtual”. “Since the interior state of another being can only be understood through interaction, no objective answer is possible to the question of when an “it” becomes a “who”, but for many people, neural nets running on computers are likely to cross this threshold in the very near future,” Blaise Aguera y Arcas, Vice President and Fellow at Google Research, Lemoine’s colleague and companion, wrote just a couple of months prior to Lemoine’s firing. “Perhaps, when it comes to AI, many of us will ultimately conclude that relationships matter more than the interior and imponderable question of “realness”.  

The process of online self-directed learning is about nourishing relationships, with our parents and partners who facilitate and accommodate that learning, with our mentors, peers and the emerging AI models who guide us to new depths of mastery and reality, but most importantly, with our own selves. Micromanaging or otherwise intruding in this delicate process to force it into an arbitrary linear path in disregard of the dynamic nature of the human mind is like disrupting a delicate ecosystem to make sure it fits a crude industrialized schematic. 

In the digital age, the web is not just something we use. Partially, it’s already our habitat as a species and it is definitely not confined to screens. As AR develops, most things we use in our daily lives will no longer need to be physical and the notorious “screens” – TV screens, computer screens, phones – will become an artifact of the past. The web as an intangible interconnected pool of information is where we learn, acquire new skills, where we earn money and start relationships. Just like any habitat, it has its own inherent dangers, but avoiding those dangers comes with masting to navigate the habitat, not with abstaining from it.

What will be the practical application of restricting children and teens from engaging in deep conversations with one another as well as with large language models or from generating immersive VR and AR worlds from imagination, because, like the overly concerned legislators tell us, they may encounter content posing “a risk to physical and mental health of a minor”? Will teenagers be restricted from having conversations about anorexia, death, sexuality? In Russia, where I originally come from, they have decided that it’s harmful and hence illegal to expose anyone under 18 to any information on homosexuality, while content glorifying war and brutalizing neighboring countries seems to be encouraged. As I am writing this, Russian authorities are pushing new legislation, aiming to ban information on homosexuality for any audience. It’s just one example of how surveillance and control regulations normalized for minors can spread to cover broader demographics.

Setting up free-to-browse hubs for kids and teenagers who don’t otherwise have unsupervised access to a computer and to the web can make a huge difference, as demonstrated in many experiments, most notably, Sugata Mitra’s Hole in the Wall. Mitra originally discovered the power of self-organized learning after he literally broke through the wall of the corporate research facility in New Delhi where he was teaching into the neighboring slum where the local kids were playing and filled the hole with a computer. In the months that followed, he observed the largely illiterate children, who neither spoke nor read English, successfully teach themselves basic computer literacy.  Over the following two decades, Mitra repeated his experiments and did extensive research both in India and abroad, through many cultures, including some of the world’s poorest communities, making computers available to children in public places, on their own terms. Each time, the children were able to develop deep learning by teaching themselves. Mitra also discovered that self-organized usage of computers and the web by groups of children can be especially empowering as an antidote to indoctrination

Mitra called the method he had developed Minimally Invasive Education (MIE), which later also became known as SOLES (Self Organized Learning Environments), recreated in Schools in the Cloud.  In a recent talk, he recalled: 

Inside classrooms across England, the US and Spain, teachers would say, ‘How did the kids get there? Shall I stand by and watch every move?’ And I would say to them, “the moment you do that, they won’t get there.”

— Sugata Mitra

This reminds me of my own recent entry in our homeschooling blog: 

I will be trying to continue capturing snapshots of their beautiful learning processes, but I wish my interference were minimal. Sometimes I get the feeling I am perturbing the course of their delicate thought and learning process by observing and capturing it for my records, just like an observation or a measurement can collapse a wave function in quantum mechanics.”

I have observed (not only in my own family but also in the greater self-directed learning community I have been following across the globe for years) that once you give children enough spacetime to explore online following their micro-motives, yourself as their caretaker always remaining their accessible go-to person for any queries, curious and open-minded about their interests, facilitating their further explorations but otherwise allowing the children to take the lead, you will be amazed how good they become with tech and how well they learn to verify their sources. You will also see them develop an expertise, a sense for what really matters and is important for them to learn — something they will be thrilled to teach you, too, as long as you remain a good listener. It’s a very humbling experience.

One of the nice perks of this lifestyle is that it’s a way to stay young at heart and retain brain plasticity, reinventing ourselves, our relationships with our children and the fluid environments around us as we move forward.