Loura Kruger-Zwart

We've been witnessing a fascinating trend in the digital landscape: the rise of casual mini-games across a wide array of online platforms. This trend isn't just a passing moment: it has truly reshaped how we interact with websites and apps, and it's changing the way companies engage with their users and keep them engaged.

The New York Times: from paper crosswords to mobile Wordle 

One of the most visible and well-known examples of the mini-game ascent comes from the New York Times. NYT has long been known for its iconic crossword puzzles, first published in 1942, which set the stage for daily brain-teasers. Its enduring popularity demonstrates the long-standing appeal of quick, daily mental challenges, and their foray into modern mini-games has catapulted them into the online spotlight. The trend started before the pandemic already—the daily Mini Crossword and Spelling Bee are global favorites—but it was the acquisition of Wordle in 2022 that truly put NYT games on a new scale. This game was unique and the results easy to share, spoiler-free. NYT’s casual game offerings have only continued to grow in popularity since then.

The trend isn't limited to news outlets. Professional networking site LinkedIn has introduced a few casual games that refresh daily, and even e-commerce platforms like bol.com have incorporated games into their apps. The OEC, an economic data platform, has brought in regular visitors with their games like Tradle and Connectrade (homages to NYT’s Wordle and Connections). The spinoffs don’t stop there: Worldle for the geographers-at-heart, Heardle for the more musically inclined, Loldle for the gamers, Waffle for our smartest friends, and Quorlde for the eccentrics among us—just to name a few.

They're quick, tricky enough to feel challenging but not so difficult as to cause frustration, and effortlessly shareable. With usually only one new puzzle per day, the process of completing them provides a sense of downtime without feeling unproductive. Everyone attempts the same puzzle and there are no ‘levels’; the results—your score—can be transformed and emoji-fied to make for spoiler-free comparison among family, friends, or colleagues. This limited availability and global consistency creates a shared experience among players, fostering both a sense of community and a dose of friendly competition.

Of course, there is nothing inherently wrong with playing and enjoying mini-games, but the context of their presence raises some questions. The NYT claims these games drive traffic to their platform and provide levity from a heavy news cycle, but they've also created a separate app for their games, which effectively results in less traffic to their site and main platform. This widespread adoption begs the question: where did this trend come from, and where is it going?

From the early days of flash games to the massive gaming industry we see today, playing games on the internet has always been a core part of the online experience.

However, what we're seeing now is different. These new mini-games provide quick dopamine hits, often taking just 30 seconds or two minutes to play. They're designed for short attention spans and busy lifestyles, often integrated into apps we use daily. The rise of mini-games and its corporate encouragement isn't accidental: it's a response to changing user behaviors and platform needs. With overall engagement declining and users spending less time on individual sites (besides the big social media platforms), these games serve as a way to increase user return, retention, and time spent.

But is this trend a solution or part of a broader problem?

A little deep dive into games online: the case of Indonesia

We took to Dataprovider.com’s Search Engine to see what we could find about how games are mentioned online. We can see a general rise in “games” being referred to across the internet, but there’s one trend line in particular that we couldn’t ignore: Indonesian websites and websites based in Indonesia have seen a meteoric rise in the mentions of games online since 2021. In Figure 1, we can see that Indonesia’s line starts rising at the end of 2021 and continues steadily upwards until the end of 2023. Then, from the beginning of 2024 until now, there’s been an exponential growth in “games” mentions that overtakes the numbers of the UK and Canada—even combined. We tested our data for an increase in English-language Indonesian websites, and discovered that no change was present that would account for the rise. No changes in our internal methodologies could explain the changes either, which sent us to the news to investigate. 

Two unique explanations came to our attention; two answers that seemed on opposite sides of the same coin. On one side of the coin, we have to take a look at gambling—an activity that is illegal for both locals and tourists in Indonesia for religious reasons. However, online gambling has become a problem in the country: in lieu of (legal) traditional casinos, the gambling that happens tends to be in the form of mobile apps and sites, where users can play gambling ‘games’ with ‘real stakes.’ These platforms create considerable political and social concern about gambling gamification, and a governmental task force was set up earlier in 2024 to address the issue. For better or worse, this helps explain the prevalence of “games” on the Indonesian internet. 

The other side of the coin paints a much happier picture, giving context to the data we see in Figure 1. It starts with the “Perpres 19/2024”, a presidential regulation and government initiative for supporting Indonesian mobile/video game developers. The Perpres was announced in 2021—aligning with when the orange trend line starts to rise—and more actively enacted in 2024, essentially when the exponential growth begins. With a large and active online (gaming) community in Indonesia, the initiative is intended to keep more gaming revenue in the country by supporting talent development, funding indie game development studios, and more. Some of the most popular games from Indonesian studios are known around the world (Coffee Talk and Coral Island, especially!). From this angle, the increased mention of “games” on local websites can be clearly attributed to this giant boost to Indonesia’s gaming industry. 

Neither of these explanations are the same kind of mini-game as what the NYT offers, but this case tells us plenty about the way people use the internet today: to find games, to play games, and to discuss games in blogs, articles, and comment sections across the world. At the same time, they all add together as a way to keep people and their activities firmly online. 

It appears that (mini-)games are the newest addition to the attention economy. Just as the printers and publishers before them, online platforms are competing for users' limited time, focus, and screen space. The seemingly harmless venture into casual games—especially from news providers—can spur a sense of service dilution, where platforms expand too far beyond their core offerings, just to keep users engaged. 

It risks becoming a symptom of content collapse, a modern phenomenon where all information flattened and given more or less the same importance. For example, it occurs when heavy news is placed alongside clickbait and quizzes, which can lead to a kind of numbing effect: there’s less emotional room to deal with dark themes, and the joy of puzzling is damped by negative events.

As we look to the future, several questions emerge:

Will mini-games become a standard feature across all types of digital platforms?

How will these games evolve to maintain user interest and generate income for their hosts? 

What impact will this trend have on user behavior and expectations?

How will platforms balance their core services with these additional engagement tools?

The rise of mini-games online represents a significant shift in how we interact with digital platforms. In our article on GenAI Transparency, we explored the impact of the ‘time spent’ metric (and how it interests shareholders): it's not hard to imagine that the interest in keeping users on the platform will take precedence over the reason the users are on the platform—as long as money is being made, and the metrics show growth, quantity will remain above quality. 

But that’s the cynical take, and it could just as well be that mini-games will stay casual, just a fun addition and well-deserved source of levity online. They offer benefits in terms of engagement and user retention on the corporate level, and in terms of a brain-break for users, but of course, casual mini-games also raise important questions about the future of online content and user experience. For those of us on the player side of mini-games, it comes down to being intentional and paying attention to the context in which the games are. It’ll be fascinating to watch how they shape the digital landscape in the years to come.

This article is part of the Modern Internet series, where we dive into today's digital culture and the online landscape. Check out Internet Nostalgia for a head start on finding great spaces to hang out online, or read our piece on GenAI Transparency to learn about maintaining the Living Internet. If you’re curious about the importance of entropy to security, take a look at Low-Tech Internet (and enjoy the lava lamps). 

Tech

Crosswords, brain-teasers, puzzles, and more: it’s hardly a surprise to find a mini-game or two, tucked away in news and social media platforms. But what does this mean for how we engage with the modern internet? And why are games so popular on the Indonesian internet, in particular? Take a look at our latest blog for insights about mini-games, modern engagement, and the attention economy.

Mini-games and modern engagement

Web trends

Culture

Tech news

Veronika Vilgis

The .ai domain has experienced an unprecedented surge in popularity, with the past few years seeing registrations skyrocket. Initially affiliated with the small Caribbean island of Anguilla, the .ai extension has become a highly sought-after digital asset, largely due to its association with artificial intelligence. Tech companies, startups, and even large corporations have rushed to secure their slice of this internet real estate, making it a go-to choice for AI-related branding. We look into the data to see what’s really behind .ai domains and if the rise is just a hype.

The growing interest in artificial intelligence has elevated the .ai domain from a relatively obscure country-code top-level domain (ccTLD) to a trendy choice for tech companies worldwide. Companies like OpenAI and Stability.ai have embraced the extension, making it synonymous with cutting-edge AI innovation. Just as investors flocked to AI stocks, domain investors have rushed to secure .ai domains, viewing them as the domain market’s next big thing. Eager to capitalize on potential resale opportunities and future value appreciation, they have snapped up as many premium keywords as possible, leading even generic or non-tech terms to be registered with “.ai.”

While the surge in .ai domain registrations is impressive, the quality and usage of these domains tell a different story. As of October 2024, there are 204,700 active .ai websites, including online stores, blogs, and parked domains. A closer analysis reveals that placeholders and parked domains dominate the landscape, making up 61% of all active sites. Business and e-commerce websites account for 23%, while content-focused sites make up 11%. The remaining share consists of low-content sites, frame sites, forums, and blogs.

To understand the meteoric rise shown in Figure 1, we examined whether the total increase is primarily driven by increased placeholders. Our analysis found that between October 2023 and September 2024, the number of placeholder .ai domains surged by an astonishing 2,271% (Figure 2). The majority (88%) of these placeholders are managed by registrars affiliated with Team Internet Group PLC (formerly CentralNic Group PLC), including 1api.net, Centralnic Canada Inc., and Hexonet Services Inc. Furthermore, 92.3% of the placeholders are hosted by Amazon.

The influx of parked and placeholder domains comes with drawbacks. For one, it can create a misleading perception of activity within the .ai domain space. Although registration numbers are high, many of these domains do not provide value in terms of content or active use, which can make it harder for genuine startups to find available and relevant domain names. As our previous analysis of .com placeholders showed, many of them have a short lifespan.

Additionally, the practice of domain parking drives up the cost of acquiring .ai domains, as owners of premium domains often set high resale prices. For cash-strapped startups, this presents a significant barrier to obtaining a domain that truly fits their brand.

Growth in Business and E-Commerce Websites

While the growth rate for business and e-commerce websites is lower compared to placeholders, it remains notable, with increases of 477% and 734% over the past 12 months, respectively (Figure 2). These websites are predominantly located in the United States, followed by Germany, India, the United Kingdom, Canada, and the Netherlands (Figure 3).

Next, let’s take a look at the topics associated with the different business, e-commerce, and content websites of the .ai top-level domain (Figure 4). The “Software & Data” category unsurprisingly dominates nearly half (44.9%) of all topics, confirming that software development and data analysis are central to the AI conversation. “Marketing & Advertisement” follows as a distant second at 14%, indicating AI's growing role in transforming digital marketing strategies. Other notable sectors include “Telecom” (7.4%) and “Entertainment” (4.6%). Topics like “Financial” services (4.4%), “Art & Design” (4.1%), and “Science” (3%) highlight AI's diverse applications across various fields, though they trail behind the more dominant themes. Meanwhile, “Health & Medical” (2.5%) and “Business & Careers” (2.2%) represent emerging areas where AI is making significant strides, albeit with a smaller share of the content focus.

With the entry of large language models (LLMs) and generative tools into the mass market, AI has been widely hailed as groundbreaking and revolutionary over the past year—and the surge in .ai domain registrations reflects this excitement. However, most of these websites are placeholders or parked domains, often created solely for resale, essentially just occupying space until replaced by something more substantial. As we have shown in a previous analysis of .com placeholders, many tend to disappear quickly. For the remaining business and content-driven websites, it remains to be seen how stable the trend will be. As the initial hype cools and doubts emerge about the true capabilities of AI, it’ll be worth watching how demand for the .ai top-level domain evolves in the coming months and years.

Ultimately, the .ai domain's future success will depend not just on the number of registered domains, but on how many of these domains are actively contributing to the advancement of AI and technology in meaningful ways. For startups and businesses in the AI space, building an active online presence with a .ai domain could offer a significant competitive advantage—if they choose to go beyond just reserving a name.

Domains

We’re seeing AI in every nook and cranny online, and the realm of top-level domains is no different. The amount of active .ai websites has skyrocketed in the last year, and now they’re everywhere—right? We dive into the data to see what’s really going on in the .ai domain space.

The .AI domain boom: what the websites behind the trend reveal

TLDs

domains

The .eu TLD is managed by EURid, the European Registry for Internet Domains. It was established to provide a unique European identity on the internet and is available to residents, organizations, and businesses based within the European Union, as well as in Iceland, Liechtenstein, and Norway. This is monitored by the registrar—the organization that facilitates connecting a chosen domain name to a chosen TLD—who collects information about the registration depending on the TLD restrictions. For the .eu case, registrar GoDaddy requires address validation at registration:

This already shows a point of contention: the registrars say upfront that UK addresses are not allowed. But how did this happen, and where do we go from here?

The UK's official withdrawal from the European Union on January 31, 2020—popularly known as Brexit—set in motion a series of changes for .eu domain holders based in the UK. These changes were implemented in stages:

From January 1, 2021, UK residents and organizations became ineligible to register new .eu domains. This swift effect of Brexit closed off a previously available option for UK entities seeking to establish a European online presence.

On January 1, 2021, existing .eu domains registered to UK addresses were suspended. This meant that while the domains still existed, they ceased to function, potentially causing significant disruption to businesses and individuals relying on these domains for their online presence. The idea was that most UK-based .eu domains would transfer before this date, but for various reasons, this was not universally possible or accomplished. 

Following the suspension period, all suspended .eu domains registered to UK addresses were withdrawn on January 1, 2022. These domains became available for general registration three months later, on April 1, 2022, for registrants with a validated EU address. This final step effectively removed these domains from UK registrants, unless they had taken specific actions to maintain eligibility.

What happened next? Consequences for UK-based domain registrants

Losing a domain can lead to various complications for businesses. Legal and contractual issues may arise when legal documents reference the domain, necessitating updates and potential renegotiations. For businesses that use the domain for their email addresses and communication channels, the loss of the domain may cause major interruptions and loss of contacts or conversations. Equally, if a domain owner has spent time and money on search engine optimization and marketing efforts tied to a domain, this effort may be lost and requires a complete rebuild of an online presence. Branding and recognizability can also suffer from a forced switch like this: domain owners may lose their online identity having to trade for a less catchy domain name with another TLD, especially if their original name is not available. It’s easy to see that ‘vanman.eu’, for example, sounds a lot more authentic and authoritative than aberdeenvanman.com.

To maintain their .eu domains, UK registrants were pushed to take one of two actions. 

Establish EU citizenship: registrants would be required to change and provide evidence of residency or citizenship in another EU member states. 

Transfer ownership: UK registrants could transfer ownership of their .eu domains to EU-based subsidiaries or partners.

The case of “leave.eu”, though the domain is no longer active, provides a notable example of the Brexit impact. This pro-Brexit campaign website, ironically, faced the loss of its domain due to Brexit. To maintain their online address, the organization elected to move its registration address to EU-member Ireland—demonstrating the lengths to which some entities (even those who desperately fought for these changes) went to preserve their .eu presence.

For this article, we were not specifically interested in the ones that maintained their domain via the mitigation strategies we mentioned above, but rather those that moved their domain to a new TLD. Where did they go, and were they able to retain their original name?

In the course of our research, we looked at 4,160 active .eu website that were allocated to a UK registrant in December 2020. As you can see in Figure 1, there’s a sharp drop in the returned response—meaning this number of websites are no longer accessible after January 2021. This was obviously expected, given the registry’s post-Brexit policy. Over time, the number continues to drop gradually. The number of UK-registered .eu websites that redirect is 828, and Figure 2 shows the top TLDs these domains redirect to. The majority redirects to a .com domain, followed by .co.uk and then to (another) .eu. Notable is that 46% of the .eu to .com redirecting domains actually redirect to www.sedo.com, a domain parking company,

Next, we were interested to see if the ones that redirect were able to maintain their name. Indeed, we found 276 domains that redirect to the same name, i.e. only the TLD portion changed. The remainders redirect to domains that are not identical to the original, but maintain part of the former name. For examples, see Figure 4. We suspect that in most of these cases the intended, ideal, and original name was not available for registration. In all of these cases, the domain name length increased; common variations included the addition of geographical location, hyphens, or company legal structures (e.g., ltd).

The Brexit-induced changes to .eu domain eligibility have had far-reaching consequences for UK-based registrants. While some were able to adapt by transferring registration to an EU-based country, many faced the loss of their established online identities. This situation underscores the importance of considering geopolitical factors in domain registration strategies and highlights the need for businesses to have contingency plans for their online presence.

The .eu top-level domain (TLD) was introduced in 2005, and has been a top choice for businesses and individuals across the European Union since then. However, the United Kingdom's departure from the EU has had significant consequences for UK-based .eu domain registrants. Can you have a .eu domain without being in the EU? Well, not technically…

Here, there, and everywhere: tracking UK-owned .eu domains in a post-brexit world

Policy-making

Brand protection

Despite feeling like the behind-the-scenes web is inaccessible to most, you’ll be pleased to know that the internet is not magic: there’s a reason and explanation for everything, and while technical, it’s not always as complicated as it might seem. One thing that brings the hypercomplex aura of the internet back down to earth is the quirky, low-tech side of what makes it work—there really are some things computers and algorithms just cannot do. 

Let’s consider randomness: as human beings, with complex thought patterns and a penchant for unpredictability, we can be random. We could pick a number between 1 and 1,000 for no reason besides it feeling ‘random,’ and follow that up with another entirely random number, so that even we ourselves don’t know which choice we’ll make next. That said, humans are not flawless randomizers, as we can quickly descend into creating patterns or memorable combinations. Computers, on the other hand, cannot be random—it is genuinely impossible. They can simulate randomness, of course, like an online dice roller or random number generator does, and it can seem more or less random to us as the receiver of information. 

This randomness-aversion also appears in other ways. When was the last time you felt your Spotify playlist was well and truly ‘shuffled’?  While it’s true that the platform’s ‘automix’ features give preference to similar sounding songs and your recent listening history, some songs will never come up because there is often no ‘true shuffle’ on algorithm-based platforms like Spotify. This is called psuedorandomness, and yes, for the most part, it is ‘random enough’ for the purposes we need on the user-end. But the computer will always know what comes next: the sequence of results is always calculated, always predetermined. 

Why is randomness important for security?

For low-stakes things like song play orders or random number generators, the lack of true randomness is not a big deal. As long as it seems random to us as humans, the mission has been accomplished.

The real issue arises when considering (online) security and computer to computer (and other M2M) communication. Much of our online experience is encrypted: consider the padlock in your browser bar, indicating SSL status, or how much time you spend ‘logged in’ on social media, email, or e-commerce sites. And much of our online experience must be encrypted to keep us safe as we share personal data like contact information, photographs, payment details, and other confidential kinds of records. Online security is based on keys, which rely on randomness and undecipherability to ensure unpredictability—if a computer and malicious actor can predict or decipher the key, it cannot be secure. Strong, unique numbers are crucial for this, as they become the secure encryption keys for wifi passwords, SSH and SSL certificates, and more.

Let’s take a look at the data: the Secure Hash Algorithm (SHA) in SSL certificates refers to cryptographic hash functions used in the digital signatures (keys) of these certificates. The functions are essential for ensuring data integrity and authenticity in SSL/TLS communications. In essence, generating an SHA hash involves processing the input data—like a prime or truly random number—through a series of mathematical operations—like multiplying by another prime or random number—to produce an output that uniquely, securely represents the original input. Figure 1 shows the four more-used SHA types for .com domains in our database.


So, if humans are not effective randomizers, and computers cannot be random, how does online security work? This is where the low-tech internet shines: it utilizes the true randomness of the outside world and the physical things around us to ‘seed’ randomness. An excellent example of this is IT management services company Cloudflare and their lava lamps. Referring to their lamp wall, cryptographer Nick Sullivan explains that it's the photographing of the lava bubbles, the room’s light ambience, (white) noise levels, and a number of other—completely offline—factors that form the basis for solid internet encryption capabilities. Their branch in London uses a ‘chaotic pendulum,’ and the Singapore office tracks the decay of radioactive isotopes to derive randomness for encryption keys. Digital security companies use multiple sources and layers to introduce randomness in their security.  

The mentioned examples are physical sources of entropy, which means they produce randomness in a physically observable manner. Other sources of entropy are just as useful. Thermal noise can be tracked to harness randomness from the hums and jitters in electronic circuits, and Quantum Random Number Generators (QRNGs) are devices that leverage quantum phenomena, such as the behavior of photons, to produce truly random numbers. When operating systems collect entropy from various sources (like mouse movements, keyboard timings, or system events) this is called Software-Based Randomness, which relates to the randomness that can be generated by human-device interaction. User-generated entropy is also a popular contribution to encryption, as Tom Scott explains: personal computers can use the milliseconds between clicks, or smartphones can use movement and tilt data to generate randomness. 

These methods highlight the importance of both physical and digital randomness in maintaining security protocols across the internet. Each method has both strengths and weaknesses, and it’s often a combination of these techniques that get employed to ensure high-quality randomness. These low-tech solutions solve incredibly high-tech problems, and help us understand how the internet works just a little better. ‘Low-tech’ should not imply that the solutions are simple or effortless, but rather that they are graspable for those outside the industry, too.


Low-tech solutions for high-tech problems

Barcelona, ES: this solar internet project is a solar-powered website and magazine, created by Belgian author Kris de Decker. It focuses on and explores the sustainability of technology, reduction of (power) consumption, and prioritization of low-tech solutions to socially constructed high-tech problems. Its solar base is in Spain, where the good weather helps to keep it online, but the site gives its visitors the heads-up that it may be down when the weather has been bad. Take a look at the website to learn about dithered images, technological sustainability, using history to inform our futures. 

This article is part of the Modern Internet series, where we dive into today's digital culture and the online landscape. Check out Internet Nostalgia, for a head start on finding great spaces to hang out online, or GenAI Transparency to learn about maintaining the Living Internet. Keep an eye out for our upcoming piece on the rise of mini-games where they're least expected, and later, the consequences of digital hoarding.

Everything to do with the internet is entirely technical, complex, and shrouded in esoteric jargon—right? Sometimes it seems like you have to be an expert to understand what’s happening out there. While there are plenty of ultramodern things about the web, the low-tech internet is what helps keep us safe on our digital escapades. Read on and dive into the high-tech (security) problems solved with 'low-tech' solutions.

How the low-tech internet tackles high-tech threats

Cybersecurity

Security

Navigating the digital landscape is hard enough, let alone introducing the drama and politics of domain names. Throw in nasty behavior and some lawsuits, and you’ve got an introduction to the world of cybersquatting.

Going beyond just simple domain registration, cybersquatting is a complex and evolving practice in the digital landscape. At its core, it involves the bad faith use or registration of domain names, often with the intent to deprive rightful owners or profit from others' trademarks. This practice has expanded to include username- and handle-squatting across online and social media platforms, extending its reach beyond just domain names.

Digital troublemakers have developed a range of techniques and forms of cybersquatting over the years. The traditional form, domain squatting, involves the registration of domain names similar to established brands or trademarks—the registration of netflix-movies.com would be an example. The practice has evolved to include username and handle squatting on social media, meaning that bad actors claim the names resembling known brands or public personalities. Typosquatting, another common technique, involves registering domains with common misspellings of popular websites—think netflux.com or netfllix.com. You can see these in action in Figure 1: there are many ways to manipulate a domain name for cybersquatting purposes.  

Internationalized Domain Name (IDN) cybersquatting utilizes visually similar characters from different scripts to create deceptive domains—netflíx.com looks just like the true domain until you notice the accent on the ‘i’. In Figure 2 we see several real examples of IDN cybersquatting, extracted from the Dataprovider.com database.

The motivations behind cybersquatting are primarily financial or malicious. Many cybersquatters attempt to sell domains to trademark owners at inflated prices, seeking financial gain, while others use these domains for more nefarious purposes, like phishing, scams, malware distribution, or just redirecting to competitor websites. In some cases, cybersquatting is used for political leverage: individuals register domains related to political figures or movements, as seen with cases from Australian senators to US presidential tickets, for negative consequences. These activities create a complex and often problematic ecosystem, affecting businesses, individuals, and the overall integrity of the internet.

Cybersquatting extends across various languages, exploits different top-level domains—like by registering netflix.jp before the company itself can—and employs sophisticated techniques to bypass traditional safeguards. It is absolutely not limited by language or geography. The introduction of new generic TLDs (like netflix.film) has added another layer of complexity to the cybersquatting landscape, both in terms of new opportunities for squatters and new challenges for brand protection. While most cybersquatting is profit-driven, there are non-commercial forms such as cybergriping, where individuals register domain names to criticize or parody organizations: ex-South Park writer Tony Morton is (in)famous for creating spoof websites and using humor to combat hate. However, this raises questions about the balance between free speech and name or brand protection.

At Dataprovider.com, we developed our own system to identify cybersquatting for the top 100,000 websites, based on the traffic they receive and a few other indicators.

Take a look at the number of cybersquatting domains we detect for each of the ten target domains in Figure 3. According to our data, nearly 2,400 domains targeted the main domain of the US based airline United in September 2024, while YouTube.com and 3m.com were targeted by 1,040 and 1,014 cybersquatters, respectively. We also see three major adult sites among the top 10: explicit links are popular targets for cybersquatters because they attract large volumes of traffic, making mistyped or similar domains highly profitable through ads, affiliate links, or redirects. Additionally, adult sites generally face weaker brand protection, allowing cybersquatters to exploit misspelled URLs and monetize them with less resistance from site owners.

Understanding cybersquatting means understanding the domain ecosystem. Key players include ICANN (the Internet Corporation for Assigned Names and Numbers), which governs the domain name system globally, registries that manage top-level domains, and registrars that facilitate the registration of domain names. This ecosystem establishes the backdrop against which cybersquatting occurs. and the foundation on which it can be combatted. 

Legal proceedings, often through the UN’s World Intellectual Property Organization (WIPO), have become a common recourse for companies and individuals fighting cybersquatting. Recent high-profile cases illustrate the complexity of these battles: Canva recently secured 174 domain names through legal proceedings, demonstrating the scale of cybersquatting targeting major brands. Conversely, the case of GigPig highlights the potential for abuse of anti-cybersquatting measures, with the company being accused of reverse domain hijacking. These cases underscore the intricate nature of cybersquatting disputes and the challenges in fairly resolving them.

New elements of brand protection have emerged in response to the threat of (commercial) cybersquatting. Preemptive domain ‘warehousing’ is a common response to these kinds of threats: companies register as many domain variations as they can think of to protect their brand. This proactive approach, while effective, contributes to the saturation of the internet's domain space.

Besides proactive domain management, AI and machine-learning tracking technologies are being developed to help detect cybersquatting. Digital tools and databases—like what we've developed at Dataprovider.com—are solid resources for monitoring particular domain names and identifying cybersquatting attempts. Many businesses implement redirection networks and systems that redirect traffic from similar domains to the legitimate website. There are now more ways to try and mitigate the issue, but it's what happens after the fact that remains an issue. On the legal front, companies and policymakers are utilizing existing laws and pushing for new legislation to address emerging forms of cybersquatting—but it remains a murky and complicated realm. 

As the internet continues to evolve, so too will the challenges and solutions related to cybersquatting. We can expect to see enhanced tracking technologies emerge, offering more sophisticated tools to identify and prevent cybersquatting attempts in real-time. The legal landscape is likely to expand, with the development of more comprehensive international legal frameworks to address cybersquatting across jurisdictions. Maintaining the delicate balance between robust brand protection measures for business, while preserving the principles of fair use and free speech in the digital space, promises to remain a hard task. With the growing popularity of emoji domains and TLD developments, we'll never be rid of the risk, but it's not impossible to keep on top of it. This digital ecosystem challenge will requite ongoing vigilance, technological innovation, and legal adaptations to address its evolving nature and impact on businesses, individuals, and the integrity of the internet as a whole.

Brand Protection

Bad manners, grudges, and name-calling… sounds like playground antics, but this is the online world of cybersquatting. Including domain squatting and typosquatting, this illicit digital activity is usually for malicious, political, or financial gain but is a popular stunt among satirists too. Read on to learn more about cybersquatting and the efforts to combat it.

What is cybersquatting? Preventing domain hijacking and brand misuse

The modern history of emoji domain names began in 2001, when the first emoji domains—☮️.com, ♨️.com, and ♨️.net—were registered. These domains were made possible by the use of Punycode, which converts Unicode characters like emojis into ASCII, allowing them to function within the Domain Name System (DNS).

While these were just a novelty at first, emoji domain names gained real traction around 2015 as mobile internet usage soared and emojis became a more central part of everyday digital communication. Emoji domain names are managed by only a few registries, and the most frequently used are .ws (Western Samoa’s country code top-level domain), .to (the ccTLD is assigned to the Kingdom of Tonga) and .fm (the ccTLD of the Federated States of Micronesia).

Emoji domain names provide an opportunity for creative and memorable branding. Budweiser, for example, registered the domain 🍺.ws, cleverly aligning its product with the universally recognized beer mug emoji. Similarly, Coca-Cola previously launched an innovative marketing campaign by registering all variations of the happy face emoji: a creative example of how to use emoji domains as a fun tool for modern marketing.

But just how many emoji domains are out there? In our database, we detect 8,262 domains with an emoji in the main name—however, the vast majority have no active website associated. As of September 2024, only 423 emoji domains have a working website, and of these, 27% are websites that provide informational content and 5% are associated with a business or company (see Figure 1). The remaining websites are placeholders, parked domains of low content websites, which typically only have one page and little overall information. Close to 1,600 emoji domains are redirects, while the rest—a large majority—return no response when queried. These low numbers suggest that emoji domains are definitely not a trend (yet), but it also means that plenty of prime domains are still up for grabs and likely to be reasonably affordable. You can check the availability and prices of emoji domains with this handy tool at i❤️.ws.

Next, let's take a look at the top-level domains (TLDs) associated with emoji domains. Only three registries actively allow registering domains with any emoji: .fm, .to and .ws. Having said that, there are certain types of symbols that can be classified as emojis, such as © or ®, that are allowed as part of other TLD registration policies. Figure 2 shows the top 10 TLDs (by number of emoji domains) that feature at least one emoji in the domain name.

Indeed, ©, ® and ™ are the top 3 most frequently occurring when we look at emojis used in .com domains. Many of you might say these are not ‘real’ emojis, but don’t fret! We’ll give you the top 20 most frequently occurring real emojis across all TLDs in Figure 3.

Our favorite red emoji tops the ranks, which is both reassuring and perhaps a little surprising. If you’ve lost faith in humanity, it’s nice to know that the internet still has heart(s). Perhaps this would be more reassuring if the laptop emoji wasn’t ranked second—the machines are catching up. 

Some emojis listed in the top 20 make sense and are in agreement with observations reported elsewhere, such as the fire or rocket emoji, and the notorious eggplant emoji being popular is not unexpected. Less obvious are the bikini emoji or the bride with a veil, but perhaps these (and a few others) hint at a certain adult-themed trend among emoji domains.

Curiously, we didn’t find a single emoji face in the top 20: no 🙂, 😆, 😂, or even a 😡, which is contrary to what we’d expect given their frequent usage in short messages or social media posts.  

Digital creatives are already making use of emoji domains to push the boundaries of the ‘usual’ internet. Take full-stack developer Ben Stokes, for example: he decided that if emoji domains were possible, then emoji email domains should be possible too… After some tests and trials, Mailoji was born. Ben writes:

Since emoji domains transcend language barriers, providing a universally understood mode of communication, there’s a cultural impact to consider. This languageless-ness makes them particularly attractive for brands and initiatives aimed at diverse global audiences. The simplicity and directness of emojis can convey emotions and ideas more effectively than words in some cases, enhancing the appeal of emoji domains in a multicultural digital world. However, there maybe challenges across cultural and generational lines if groups interpret the same emojis differently: what looks like a friendly domain for eggplants to one person, could represent something else entirely to another. Similarly, not all emojis are displayed equally across devices and operating systems. And, the fact that emojis still need to be translated to text means they are often displayed as just a random selection of letters. For example, the domain name 🍺.ws mentioned above is converted to https://xn--xj8h.ws/—which looks a lot less appealing and trustworthy, defeating the purpose of recognizability.

Emoji domain names face certain limitations and challenges, despite their charm. For one, not all top-level domains (TLDs) support emojis, restricting their widespread adoption. Nevertheless, as technology continues to advance and internet usage becomes even more integrated into daily life, the demand for distinctive and memorable digital identities is likely to grow. It'll be interesting to see if more registries open emoji domain registrations in the future, and there's also the possibility of introducing emoji TLDs. We've written extensively about new generic TLDs: over the past years, more than 2000 new TLDs have been created, from .ninja to .dog and .yoga—there is now likely a TLD for your favorite hobby or industry. Is it just a matter of time until ICANN introduces the first emoji TLD? 

Ultimately, emoji domain names represent an intriguing intersection of technology, communication, and creativity. From their humble beginnings in the early 2000s to their current status as a novel branding tool, these domains are carving out a unique niche in the digital landscape. Emoji domains offer a playful and powerful way to stand out online, and while challenges remain, the future holds potential for further expansion and increased adoption. As emojis continue to permeate our digital interactions, the allure of emoji domain names is likely to grow and makes them an exciting area to watch in the world of internet domains.

In this ever-evolving landscape we call the internet, emoji domain names represent a quirky and captivating innovation. Combining the ubiquitous and expressive power of emoji characters with the functional utility of domain names, these digital novelties offer a unique blend of communication and technology. From their origins to their current status and future potential, let’s take a look at the fascinating world of emojis.

Emoji domains explained: a data-driven analysis

It’s both clear and inevitable that the internet is changing. The previous installment in our Modern Internet series delved into Internet Nostalgia and what’s changed in the last decade about the way we use the internet. It included some reasons behind these changes and how to recapture a bit of the magic of the ‘old internet,’ but the evolution of the online world is far from over. In just the last few years, mainstream access to and use of Artificial Intelligence (AI) has increased exponentially. AI—a term often used as a catch-all term for all things black-box and automated, here referring to Large Language Models, text-to-image generators, and machine learning algorithms—has positively influenced workflows as a tool and expanded what many have thought was possible. 

The release of ChatGPT and DALL-E preempted the takeoff of many more generative AI (genAI) tools, including Midjourney and Stable Diffusion for images, and Claude, Gemini, Copilot, or Aria for text. Many software tools have developed their own AIs for different purposes, such as Notion’s generator for longer, shorter, or simplified texts, or even the paraphrasers and summarizers of top-line spellcheckers like Grammarly or LanguageTool. 

But beyond the marvels of AI, including generative technologies, there’s something else sort of sinister happening on the internet. Bots have been an issue on social media for many years, when considering fake followers and accounts spreading propaganda or fake news, but it’s taken what feels like a new form: AI-generated content—both images and texts—shared by what seems like AI-run accounts, being interacted with by other AI-operated ‘users’. This cycle ends up looking like faceless bots creating content for faceless bots, on platforms like Facebook meant specifically for, well, people with faces. Whether the point is engagement-farming or just using sheer volume to drown out other kinds of online content, we’re left asking: what’s actually going on here?

Machine-to-machine (M2M) interactions and technologies are not new: they’ve been used in business contexts for years. Cybersecurity firms like Darktrace use AI to detect AI or other automatically generated emails, blocking them from client servers—while other AI learns this and works around it… before Darktrace figures out the workaround and blocks it… and the cycle continues. Payment terminals are essentially one machine contacting another machine (the bank’s tech) to approve and execute a transaction, and the Internet of Things (IoT) network is almost exclusively machine-to-machine communication. The web traffic from crawlers, the technology with which Google and Dataprovider.com index the internet, is classified as bot traffic, and this is also M2M tech. This is to say that there are many useful things that come out of M2M tech, but what's missing is the point of a bot-run and bot-interacted surface web.

Internet analysts have been grappling with the “Dead Internet Theory,” the idea that the internet is already or soon will be devoid of human interaction in the way we’ve been used to in the last years. Explained here by two Australian tech academics, this theory contends that “many of the accounts that engage with [AI-generated] content also appear to be managed by artificial intelligence agents. This creates a vicious cycle of artificial engagement, one that has no clear agenda and no longer involves humans at all.” The motive here is unclear, and may or may not be malicious. 

But the intention can be transparent too, with financial incentives being a major factor. Previously lively sites, like BuzzFeed, have come under fire for publishing low-effort AI content. From quizzes and listicles to travel guides, the company says “generative AI will replace the majority of static content… [and] lead to new formats that are more gamified, more personalized, and more interactive.” While generative AI can produce many times more results (from brainstorm notes to full-fledged articles) than a human worker can, and for a fraction of the cost, does this matter if people don’t actually want it? If the metric is ‘time-spent,’ as advertisers and stakeholders prefer, then AI-generated content, creating highly personalized and ‘endless’ experiences, is the right choice. 

How AI transparency can help the internet 

There are plenty of great uses for AI and genAI in the world today, from the artistic and entertainment value of creating something with these tools or enjoying the creations of others, to utilizing them to simplify your workflows. AI has become part of our online ecosystem, but the issue arises when we encounter AI, knowingly or unknowingly, where we don’t expect or appreciate it. At its core, the issue is not the existence of AI tools, but the transparency around its use. And just because it can be used doesn't mean it should… So what to do?

Detection and tagging tools are still in their infancy, but are becoming increasingly useful for spotting AI-generated content online. Social media platforms like Vimeo and Instagram have added labels to mark posts made with AI (or at least provide AI info), which creates an avenue for voluntary transparency online. LinkedIn and TikTok are expected to follow suit in the near future. These are directed at visual content, like images and videos, which leaves some questions about text-based content: how can generated written work be detected? This is much trickier, as texts are far less likely to be labeled or obviously genAI influenced. Detection applications are not accurate enough to be reliable yet, so this is still difficult to spot and even more difficult to prove. 

This means knowing your sources. Things are changing faster than we can keep up, but it’s important to recognize and critically view what we see online. Companies like Google and Adobe are already in the process of adding provenance trackers to image-based online content, the feature showing if something was taken with a camera, edited in Photoshop, or made with artificial intelligence. In the meantime, however, there’s a lot to learn about how to spot AI content, and understanding its markers can help us notice, flag, or avoid it. Trace the source of something questionable to determine its reliability, for example by finding the original poster, the artist or author, or other online mentions. And ultimately, we can avoid the online spaces that care less about AI transparency than we do. 

Written content markers: check for signs like repetition, predictability, and oddly consistent sentence length and paragraph structure. For comparison, look up other examples of writing by that author or on that topic, and see how the style and content aligns. There’s no harm in checking claims against other sources, or reaching out to the author for clarity. 

Visual content markers: there are a couple of well-known signs to look out for. Especially with images of people, you might spot hands with too many fingers or feet with too many toes, but the key is to zoom in. Are shapes in the background vague or distorted? Are cars facing the wrong direction on the road? Do lines or outlines start and stop in illogical places? Is the writing actually words, or just a jumble of lines and letters? Besides the brushed glow that genAI images tend to have, look out for things that would never appear in a photograph or video, or that (digital) artists would not purposefully include. 

Quizzes: it’s actually quite entertaining to challenge yourself on your genAI recognition skills. Start with this image quiz and get more insights about spotting AI, then follow with this video test and check your assumptions. There are plenty of quizzes around to help keep you sharp.  

If the ‘Dead Internet’ is where humans are no longer creators and interactors, the Living Internet might be where we actually want to be. At risk of repeating the conclusions of our previous article about Internet Nostalgia, there’s much to be said about the return to an ‘old’ way of using the internet—a return to online communities. User-generated and official spaces are the online environments where writers, artists, and other creators have both the incentive and accountability for sharing living content: this can even include AI-assisted works if the use is transparent or otherwise expected. There is plenty of room for responsible innovation and the avoidance of ‘bionic duckweed’. Let’s create, support, and enjoy the Living Internet together. 

This article is part of the Modern Internet series, where we dive into today's digital culture and the online landscape. Check out part 1, Internet Nostalgia, for a head start on finding great spaces to hang out online, and keep an eye out for our upcoming pieces on low-tech internet solutions and the rise of mini-games where they're least expected.

AI and its generative tools are here to stay, but that doesn’t mean it won’t take some getting used to. Its impacts are already being felt in workflow-enhancing applications and content creation capabilities, but the opacity behind its use remains troubling. What are companies and social media platforms doing to ensure integrity and credibility, and how can we as users stay sharp online? Read on as we discuss the problem of genAI transparency, explore the dead and living internet, and share ways to recognize genAI in the digital wild.

GenAI transparency and the living internet

Colin de Vries

Online advertising is an integral, though sometimes annoying, part of our digital experience. Whether you want it or not, advertising shapes the way we browse, shop, and interact online. But what might seem like just a small nuisance—think banner ads, sponsored posts, or a pre-roll video—actually relies on a complex behind-the-scenes ecosystem. This system involves a network of advertisers, publishers, ad networks, and tech platforms, all collaborating to deliver seemingly omnipresent advertisements. Understanding how these ads are crafted and targeted, and the technology used to deliver them, can shed light on the machinery that powers the digital advertising industry.

To advertise online, advertisers need to find the websites where their message can appear to specific targeted audience segments. Advertisers pay for their content to be published, with payment models based on metrics like cost-per-impression (CPM) or cost-per-click (CPC).

This is how advertisements get on websites most of the time:

These platforms collect available advertisement spaces from many sites, then match advertisers to those needs. The networks enable advertisers to manage their advertising inventory on many sites at once, meaning they don't have to negotiate directly with every single site themselves. A prime example of this strategy is working with Google AdSense.

Programmatic advertising enables advertisers to buy advertisement space in real-time with automated technology. It is an algorithm-centric approach to understanding how users interact with advertisements, and involves a system that ‘learns’ where best to place advertisements for maximum effectiveness, in real-time bidding (RTB). The Trade Desk is an example of this.

Some advertisers negotiate directly with website owners to buy advertisement spaces. This is often the case with high-traffic websites, where certain advertisement placements can be expensive but incredibly lucrative. Direct purchase advertising is common on websites like BuzzFeed, The New York Times, or CNN.

This kind of marketing and advertising involves websites that host advertisements from other businesses.  The host websites gain a commission based on the sales or leads they accrue for these ads. Examples of popular affiliate programs are Amazon Associates or, in the Netherlands, bol Affiliate Marketing.

When a user visits a site but doesn't make any purchases, advertisers collect cookies and 'schedule' advertisements for that site or product on different websites of their choice. The idea is to drive the visitor back to the original site later, after they left without making a purchase. We’ve all experienced this: being confronted with products or holidays we recently searched for, as a banner ad on a seemingly unrelated website. Facebook Ads is known to use this strategy commonly.

These advertisements are intended to match the look and feel of a site just like regular content and offer a less intrusive advertising experience. They are typically controlled through native advertising platforms specifically. On Instagram, native ads appear in users' feeds and are designed to look like regular posts, and on Amazon, native ads appear as “sponsored” products within search results.

Why do we detect online advertising technology?

Different advertising strategies require separate technology and algorithms to achieve their goals of matching the “right” ad to the intended audience. These differing demands have made the advertising technology landscape particularly complex due to its diversity. With numerous players, including many smaller, niche companies, staying up-to-date with advertising technology requires constant attention. 

Understanding fingerprints in advertising

To detect what kind of technology is being used on a website and identify the advertiser behind it, we look for what is commonly referred to as ‘fingerprints.’ A fingerprint, in the context of technologies, refers to unique identifiers embedded within the code of a website, indicating the presence of specific advertising platforms or technologies. These fingerprints can be fragments of code, URLs, cookies, or specific headers that are characteristic of certain technologies. They serve as digital signatures that help in the identification and classification of the technologies a website uses.

Let’s take a look at the fingerprints of Google AdSense. Within the below code, elements like "pagead2.googlesyndication.com" and "adsbygoogle" serve as fingerprints, indicating the presence of Google AdSense on the website.

We can take nu.nl, one of the largest commercial news sites in The Netherlands, as another example. Advertisements are prominently displayed on the front page and often overshadow the content itself. By examining the source code of the website, we can detect links to various advertising platforms. These snippets are used as a fingerprint to identify providers like Google AdSense, DPGMedia Advertising, and DoubleClick.

The challenge of manually detecting technologies

Traditionally, identifying new fingerprints for undiscovered technologies was a manual process. Our technology detection team would inspect the source code of countless websites, searching for the unique identifiers. This method, while effective, was incredibly time-consuming and resource-intensive.

Enhancing fingerprint discovery with machine learning

Being tech people at a tech company, we built a new system to improve the fingerprint discovery process. It uses machine learning to analyze large quantities of web data, detecting potential fingerprints for both known and undiscovered advertising platforms. 

This is possible with a random forest classifier, a machine learning method that utilizes the predictive power of multiple decision trees. During training, the classifier generates multiple decision trees, each of which makes its own prediction about whether a specific pattern could represent a new—in this case, advertising—fingerprint. The most commonly predicted outcome is the final classification, allowing us to confidently detect new fingerprints even when they have subtle variations.

In Figure 4, each decision tree independently classifies the data: here, a shape salad. Some trees may classify the system as “Class-A” (hexagon), while others might classify it as “Class-B” (triangle). The random forest classifier assesses these classifications through majority voting, leading to a final, more accurate prediction: “Class-A.”

Our machine learning model does the time-intensive work of preliminary identification, but our human component is the key to the system. Our technology detection team reviews the model's predictions, verifies the found fingerprint, and ensures the reliability of each new discovery added to our database. 

The initial deployment of our machine learning model has significantly improved the detection of advertising platforms that were previously undetected, broadening our understanding and offering better insights into the market. For recognized platforms like Google AdSense, our model has substantially increased detection rates. 

Advertising is a large part of everyone’s contemporary online experience, and we can learn plenty from looking behind the scenes. The technology of advertising that we detect tells us more about the active companies, preferences, and trends online. Due to the ever-evolving nature of the web, there are always new technologies—and new digital fingerprints—popping up, and machine learning tools have helped us move with the changes. The human element to technology detection is what ensures high quality, and these new tools make the resource-intensive project more efficient. Significantly, our project also lays the groundwork for other innovations: we anticipate further enhancements in our detection capabilities, allowing us to rapidly identify and adapt to emerging advertising technologies. From here, we’ll be able to provide timely insights into the latest tools and trends in digital advertising and other industries, too. We’re in the process of exploring the potential of our machine learning techniques, applying them to other fields across the web, which promises even further impacts. 

How do we know which advertisers are growing, which publishers are popular, and what ad networks have the most impact? It all comes down to Technology Detection, but this practice—while very efficient—is resource-intensive. We streamlined our fingerprint identification process with a bit of help from machine learning. Read on to learn all about how online advertising works, how we track it with algorithmic assistance, and why the human component remains central to digital work.

Advertisement detection: how machine learning changes the way we discover technologies

Technology detection

There’s a conversation going on in cyberspace, on social media and forums, in news columns and opinion pieces, about something called internet nostalgia—a longing for the ‘old internet’ and the way online life used to be. But what are we really missing? Is it the charm of slow connections and bad graphics, or having small digital communities, or just the lack of social media platforms? Is the old internet really gone, or have we just forgotten where to find it? Read on to dive into what internet nostalgia is really about, what the alternatives might be, and how to navigate today’s internet for a nostalgic (and daresay, good) time. 

Things used to be so different, but what’s so good about the old internet?

That’s certainly true, the changes are clear. The internet of the late 1990s and early 2000s had the character of an informal Wild West you could visit now and then, in contrast to the expectation of being online we face now. Not only has access to the internet been vastly expanded, an internet connection has become basically essential. It’s a necessity for many kinds of work, most kinds of entertainment, and, increasingly, for social relationships too. The rise of smartphones has been accompanied by cheap data on the go. This means we no longer have to wait to get home to go online, and we no longer necessarily sit behind a big, navigable screen to browse the internet. 

It makes sense, then, that our association with being online has changed: in the same way that a hobby without balance can become a chore, what used to be more of a down-time activity is now an integral—if not mandatory—part of our days. Regulation of advertising, business, access, and social media has replaced the ‘lawlessness’ of before. We’ve traded in anonymous and ‘purposeless’ web surfing for logged-in and targeted browsing, as we seek convenience, speed, and quick engagement. 

As with most things, there is good in the bad and bad in the good. One of the main differences between the internet then and now is how we get information.

You needed to visit a different website and different types of websites for every single thing you did online. You’d stop at a forum for advice, a business website for contact information, a maps site for directions, find a city guide, organize travel, get recipes, recommendations, games, news—these things rarely overlapped. Blogs were huge, creating communities around certain topics, and there was much less of the ‘corporate’ influence we see now, of companies running websites. Rather, the online sphere felt decentralized and much more like it was just run by individuals or interest groups. Cooking and recipes blogs might be one of the most popular blog-style internet spaces that are still often visited, and, even then, that ‘jump to recipe’ button is a quick click most of the time. While visiting all these sites and surfing around online might’ve been entertaining, getting information was a time-consuming endeavor before many of these functionalities became integrated on a few platforms. With Google Maps, for example, we can now get our maps, recommendations, reviews, and directions all in one place. 

How social media impacts digital lifestyles

It seems we’ve reached the other end of the spectrum, resulting in a new set of frustrations. One of the major complaints from internet-nostalgists is the contemporary centrality of social media. It seems like everything revolves around the big social media platforms that have changed the way we use the internet. Recommendations, discussions, opinions, purchases, entertainment, updates, social connections… does everything happen between Meta, TikTok, and Reddit now? In the earlier days of Facebook and Instagram, most users were also ‘creators,’ posting for their friends and network, whereas it's more likely now that a user simply consumes the endless-scroll of curated content from social media professionals. 

The dependency on legacy social media platforms is changing, however, as users get flooded with ads and grapple with algorithms that prioritize some content over others. Then there’s the issue of disallowed or discouraged content, and the influence of AI and bot-generated engagement. These frustrations send internet visitors looking elsewhere for the enjoyable time they used to have online.

What’s the alternative? How can we enjoy the internet again?

With the many optimizations, improvements, and modern online advents, it seems that people forget that the rest of the internet is still out there—or they don’t know how to find it. 

But, of course, the rest of the internet is still out there! Websites are easier to make than ever with specialized building and hosting platforms, whether the site is visual or text-based. Plenty of people know how or are learning to code, ready to make their own sites. Popular AI tools make this even easier, and more accessible to those who are interested. Essentially, we are not restricted to social media to find interesting and unique content, but rather there truly is space online for everyone and everything they love. 


In July 2024 alone, we counted 161 million active websites in our database. Close to 28.3 million of those are classified as content, blog, or forum websites, which means these sites exist only to inform, connect, and share. They have no obvious commercial focus, and could be the sites that come close to the internet we used to know. 


The point is to emphasize what made the ‘old internet’ entertaining and exciting: we start by bringing the focus back to community and community-generated content. This can take many forms these days, including interest-based newsletters via Substack, group chats in an app like Geneva, or active channels on platforms like Discord, Twitch, or Patreon. 


Besides this active community engagement online, there’s plenty of educational and entertainment value in seeking out the unusual websites of today. These could be passion project sites like this ecological exploration or this creative profile, unusual trackers like this interactive doorbell, or silly mini-sites like this dizzy option. Once you start visiting these delightful little corners of the internet, it’s hard to stop. But where do you find them?

How to navigate the internet for a nostalgic time

Start on the big platforms, then follow the external links of the people or topics you like. Actually go to their websites and poke around.

Search for blogs on a topic you're interested in.

Sign up for newsletters and platforms that support what you’re looking for.

Literally, anywhere: https://random.dataprovider.dev/ (developed in-house, this link that takes you to a new, random site every time you click—and don't worry, all the nasties have been filtered out)

Somewhere useless: https://theuselessweb.com 

Newsletters: https://substack.com/browse/staff-picks  | TIP: Dataprovider.com’s monthly newsletter contains a mini-round up of interesting links every month—subscribe here.

TinyAwards: https://tinyawards.net or their 2023 Roundup

The personal websites of developers & programmers

The internet still has its magic, but we’ve been a little swept away by the glamour and ease of catch-all destinations like search engine snippets and social media platforms. Treat the internet like a library: enter with an open mind (don’t head straight for the ‘new’ shelf), take your time to browse the shelves, open up a book now and then and get a taste of what’s inside, and then, if you like what you see, check out! Instead of letting yourself be served up whatever an algorithm recommends, go out into the World Wide Web and poke around until you find something fun. Don’t forget to then subscribe or check in regularly, and you’ll bring the old internet back to your life.

This article is part of the Modern Internet series, where we dive into today's digital culture and the online landscape. Keep an eye out for our upcoming pieces on GenAI transparency, low-tech internet solutions, and the rise of mini-games where they're least expected. 

Generative AI and the ever-growing reach of social media are among the elements fueling the ‘internet nostalgia’ conversation. Did we really have more fun on the internet 10, 20 years ago? Is that enjoyment a relic of the past, or do we just need to be reminded where to find it? Read our insight to learn more about this nostalgic sentiment and how to reclaim a bit of that online magic.

From slow connections to social media: exploring internet nostalgia

The domain extension .com is undoubtedly the largest and most well-known top-level domain (TLD) on the web. Operated by Verisign, the same company that runs .net, .com’s history dates back to the early days of the internet. It was one of the original TLDs established in 1985, along with .gov, .edu, .mil, .org, .net, and .arpa. It quickly became the most popular TLD due to its association with commerce and companies, most notably during the late 1990s, as businesses rushed to establish an online presence. Solidified by its namesake relation to the dot-com boom—and subsequent dot-com bubble burst as many internet-based companies collapsed in the early 2000s—.com has since become synonymous with the internet itself.

Currently, the .com TLD remains the most widely used and recognized domain extension globally. It continues to be the preferred TLD for many businesses, startups, and individuals seeking to establish a strong online presence. And, despite the introduction of numerous new gTLDs in recent years, it retains its dominance and prestige. 

But what can we learn about the internet from just this one TLD? As it turns out, plenty: the global prevalence of .com means it captures a rather interesting snapshot of the total web in comparison, giving us a glimpse into the hosting, management, technical, and content details of the online world. Time to dive in and explore what this means!

We love details, so let’s take a deeper look at the .com zone file—the text file that contains mappings between all currently registered domain names and their corresponding IP addresses. As of August 2024, there are just over 158 million registered domains in the .com zone file. In Figure 2, we see that active, available websites make up 50% of those registrations, and the remaining 50% are redirects, return an error response, or deny access to crawlers. 

Forty percent of these active websites are placeholders, 25% are business, and 6% are e-commerce websites, while the remainder comprises content pages, blogs, or framed websites. To put these numbers into context, we created a comparison set of 100,000 randomly selected active websites that are not .com. Can this comparison set show whether .com reflects overall web trends, given its global appeal and size? The random sample is made up of 65% country code TLDs, 19% generic TLDs, 16% new generic TLDs, and less than 1% sponsored TLDs. Our dataset here is made up of 32% of business and 6% e-commerce websites, though the number of placeholders is lower, at only 33%. The e-commerce trends are comparable, but .com is made up of overall more placeholders and fewer business websites. This is surprising since we think of .com as the commerce-focused TLD, but it seems like businesses overall prefer local (ccTLD) or themed new gTLDs.

The vast majority of .com domains are hosted in the USA, followed by Germany and Canada (see Figure 2). However, .com domains are hosted all over the world, from Zimbabwe to Monaco and Turkmenistan— basically every country hosts at least a few .com domains.

Those top three hosting countries are reflected in the global data, too: the US, Germany, and Canada host the most websites regardless of TLD, showing the broad appeal and impact of .com as it reflects global trends so closely. In our comparative sample, the US and Germany also rank at top two, with 41% and 14% percent respectively. Ranks 3, 4, and 5 go to France, the Netherlands, and Russia. These results aren’t unexpected, given the large proportion of ccTLDs in the comparison group.

Content management systems and social media

Next, let’s take a look at some website statistics: if we’re seeing that .com reflects overall web trends, then choice of content management system (CMS) is no exception. WordPress is the most commonly used CMS among .com websites, with 54% of them built with WordPress. WIX (12.6%) and Squarespace (10.7%) rank second and third among .com’s CMS providers. With some percentage point differences, these top three also dominate in our comparison sample, with WordPress at 66%, WIX at 7.4%, and Squarespace at 4.2%.

Tracking the popularity of social media platforms mentioned on .com sites, as seen in Figure 3, reveals that Facebook is still the dominant platform, followed closely by Instagram. X (Twitter), LinkedIn and YouTube come next, while TikTok ranks sixth with only 2.6% of .com websites mentioning the short-form content platform. As TikTok gains more traction as a business and marketing tool, we’re curious to see how these numbers change in the months and years ahead. The rest of the web has a very similar distribution of social media mentions on websites. Our comparative sample indicates that Facebook is also most common on the rest of the web, along with WhatsApp. X (Twitter) is more prevalent on .com websites than on the combination of other TLD websites. Not shown on the graph, but still noteworthy: Telegram made it into the top 10 globally with mentions on 2.2% of all websites, but did not rank in the top 10 for .com.

Average incoming links, pages, and word count

The majority of .com websites have no incoming links. For 82.5% of them, no other websites link to their address—this number aligns with the comparison sample, where 82.1% also show no incoming links. At the other extreme, 16,941 websites (less than 0.1%) have more than 1,000 incoming links, which means that over a thousand other websites link to that one (non-redirecting) website. A .com website has 18.2 pages and an overall word count of 248 words per page, on average, and these numbers include all types of websites, including placeholders. If looking only at .com websites classified as  ‘business,’ the average number of pages increases to 34 and word count to 558. In our global comparison sample, ‘business’ classified websites have 599 words and 36 pages on average, considerably more than on .com sites in general. 

Follow us

Questions?

Insights from web data

Industries

Do most online stores have a physical store as well?

Where have all the placeholders gone?

Uncovering networks of websites spreading misinformation about COVID-19

Explore more topics

Recent videos

Jorn talks Trust Grade

Jorn talks Search By Example

Feature Launch | Search By Example