Over 20 Years, Web Addresses Quietly Stopped Mattering — a Microcosm of the Internet's Drift from Equality to Big-Tech Monopoly | QDaily Business History

The rise and fall of domain names mirrors the changing gateways to the internet.

On December 7, Walmart announced that it had bought art.com. Its company blog called art.com "the world's largest online retailer of art and wall décor," though it disclosed no revenue figures.

Any company that calls itself the "world's largest" in a category buried under a string of qualifiers deserves a second look — especially now, when most people buy things on catch-all platforms like Taobao, JD.com, or Amazon. For Walmart, this was not a big deal. The company had already snapped up a string of e-commerce businesses in hopes of rescuing its floundering online retail operation; art.com was one of more than a dozen such acquisitions.

What is worth noting is art.com itself. This is plainly a rare web address — the kind people scrambled for in the internet's early days. Behind it lies the whole story of the dot-com bubble inflating and bursting, of domain names going from crucial to irrelevant, and of the internet shifting from a level playing field to a landscape ruled by a few giants.

art.com was founded in 1997, a year before both Google and Tencent, at the peak of the internet boom.

This is not a Silicon Valley startup story. When the internet became a talking point in the mid-1990s, art.com founder William Lederer was a modestly successful Chicago businessman — he ran his stepfather's wholesale art-supply business while managing Minotaur, an investment fund he had founded in 1987 and gradually grown.

In 1997, Lederer, then 35, launched art.com to sell reproductions of photographs and paintings online as home décor. It was not a particularly good business. The entire United States had only a few million internet users at the time, and people rarely swap out the art on their walls.

But at the height of the dot-com bubble, none of that mattered. What mattered was the name. Lederer paid $450,000 to buy the art.com domain from Advanced Rotorcraft Technology (ART), a helicopter engineering consultancy — about $700,000 in today's money.

It proved to be a remarkable bargain. Soon after launching, art.com raised tens of millions of dollars from the Silicon Valley venture firm Benchmark (the first to invest in Uber) and from SoftBank. Within two years, Getty — the world's largest image library — bought art.com for $115 million. At that point only 30,000 people had ever bought anything on the site, which put each user's value at $4,000.

A good domain name, a business that didn't quite hold up, and a bit of luck: that was once all it took to build an internet company and walk away with a hundred million dollars.

Now? When Google reorganized and created its parent, Alphabet, the domain Alphabet.com was already taken. Google still hasn't bought it back — and doesn't seem to care.

Domain names existed before the internet left the lab. They were once a symbol of equality.

The Domain Name System (DNS) was born in 1984, which made the world's first ".com" domain possible: Symbolics.com, registered on March 15, 1985. The DNS was created so that people could use domain names that were easy to remember (www.qdaily.com) instead of the irregular strings of digits that make up IP addresses (192.0.34.65), which are far harder to recall.

The ".com" at the far right of a domain is the "top-level domain," which signals a type of organization. ".com" means "commercial" and is now the most widely registered and universally recognized extension. ".gov" (government) is used for U.S. government sites, and ".net" (network) was originally meant for providers of network infrastructure.

The Mac may have arrived in 1984 too, but for years few companies bothered to register domains. In the first year only five did. According to Verisign, just 100 companies registered domains between 1985 and 1987 — most of them names we know well today: HP, Intel, Apple, IBM, Cisco, and Adobe.

Domains were registered in earnest only after the birth of the World Wide Web. In 1989, the British scientist Tim Berners-Lee defined the Web — in simple terms, a system of web pages, each carrying hyperlinks, connected to one another through those links.

What Berners-Lee proposed was a radically equal idea: any person or organization could have its own website and its own domain, all linked to one another.

At first Berners-Lee even kept a World Wide Web Virtual Library, a directory of every website on earth. By the end of 1993 there were only 623. They were listed in plain alphabetical order. Everyone was equal.

In 1994, with the arrival of the browser, anyone with a networked personal computer could visit websites with ease. The Web's explosive growth had begun.

As internet companies were born, the domain name was the first marketing move.

Bubble-era companies were not the only ones who craved a memorable domain. The internet companies of that era that went on to last were just as fixated on them.

Take Amazon, founded in 1995. Jeff Bezos's first choice of name was "Cadabra," from "abracadabra" — he wanted the company to grow as if by magic.

But by the time Amazon launched in 1995, the name had become amazon.com, after the world's largest river. Bezos later said he worried people would mishear "cadabra.com" as "cadaver.com." There was a more practical reason too: in those early days, when there were still few sites, many places listed recommendations alphabetically, and an "A" put you at the front. Bezos also registered Awake.com, Browse.com, and Bookmall.com — all near the top of the alphabet.

NetEase's Ding Lei missed the era when domains were there for the taking. By the time he started his company in the 1990s, every three-letter domain was gone — and letter-based domains held little appeal for Chinese users anyway.

Ding Lei's answer was to use three digits. There was no broadband then; going online meant dialing up over a phone line, and the access number was 163. So Ding Lei dialed 163 to get online, found that 163.net and 163.com were still available, and grabbed them at once. He also registered 188.net, 188.com, 126.net, and 126.com. "188" sounds like the word for "getting rich," and 126 was a common pager number at the time.

NetEase as a company owes its very existence to 163. Its earliest product was a Chinese-language email system, but it lacked the license to run the business itself. In February 1998, Ding Lei sold his email software to Guangzhou Telecom, throwing in ownership of the 163.net domain. The software was a hit: 263, State Net, 990, 371, and Zhejiang Jinhua 188 all bought NetEase's email system. The revenue from selling email helped make NetEase one of China's three big portals — and, eventually, a company listed on Nasdaq.

Flipping domains was a business of its own, beyond the entrepreneurs. Consider Rick Schwartz, known as the "Domain King." He bought his first domain in 1995 and in 2002 sold men.com for $1.32 million. After two decades of buying and selling, Schwartz now holds 7,000 domains. By his own estimate they are worth $500 million, and they include Queen.com, eTrip.com, and iHotels.com.

China's version was Cai Wensheng. The domain 265.com — recently caught up in a Chinese search-engine scandal — is the one Google bought from Cai in 2008. To this day it still carries the slogan "the computer homepage most Chinese people use," branded as "Google 265.com."

The sheer number of domains also captured the era's mania. In 1995 there were 9,005 domains on the internet; by 1998 there were 10 million. In the year after the bubble burst in 2001, a million domains were abandoned.

art.com's good luck didn't last. Less than two years after the Getty acquisition, the company was sold off cheaply. Lederer's investment firm poured the profits back into more internet companies and finally shut down in 2007.

The importance of the domain name itself began to fade with one development.

Search engines swallowed the address bar, and the internet got a single entrance.

"What do you think a browser is?"

"A place where you can search for things?" "It's a search engine, isn't it?" "Google"…

This exchange comes from a Google street-interview video posted in 2009. In most answers, the browser people used to open web pages was the same thing as a search engine — whatever you wanted to do, you started by searching.

It took Google ten years to plant that idea in the public mind.

When Google registered its domain in 1998, the world already had 10 million domains. Even if Amazon's name began with "A," finding it in a directory was no small task. Google's automated web-crawling, paired with a smarter way of ranking pages, made search far more accurate and far more pleasant to use than what came before. People now had an active way to find information.

What set Google apart was how it ranked results. The algorithm was complex, but in short, Google used backlinks: the more pages linked to a given page, the higher that page ranked for related keywords. This upended the old ideal of equal domains — the links shown to users now came in a fixed order of priority.

The Chinese internet followed. In 1999, Robin Li returned to China and founded Baidu. At first Baidu was a behind-the-scenes tool for businesses, supplying a suite of search technology to various websites. Sina, Sohu, and other portals all used it to offer Chinese-language web search. In under six months, Baidu had taken roughly 80 percent of the domestic market for website search technology.

The model copied Inktomi, a company that supplied search technology to more than a hundred U.S. portals under the slogan "Behind your e-success."

In 2001, Baidu changed course and launched an independent search engine at baidu.com, aimed straight at individual users. The next year, in 2002, it released an IE Search Companion — a search-engine plug-in embedded in Internet Explorer. Since everyone knew English domains were hard for Chinese users to remember, the plug-in let them type Chinese keywords directly into the address bar and pull up results at once.

Meanwhile Google was rising fast in the wake of the dot-com bust, with advertising as its engine. Google had put the internet in order, and advertisers naturally paid to have their sites appear beside the results. Google's advertising revenue jumped from $70 million in 2001 to $3.14 billion in 2004, the year the six-year-old company went public on Nasdaq.

Search changed how people used the internet for good. You no longer had to remember a domain, or dig through navigation sites and bookmarks to reach a particular site. You only had to know what you were looking for and search. You didn't even need to know exactly — a vague keyword would do.

In a 2002 episode of Buffy the Vampire Slayer, the character Willow asked Buffy, "Have you Googled her?" It was the first time "Google" was used as a verb on television.

After going public, Google dominated search, holding more than 50 percent of the market in 2007 — more than double the share of second-place Yahoo.

Search advertising matured into a real business. Optimizing a site's ranking on Google or Baidu lifted its position across virtually every search engine in the world. PPC — pay-per-click — let Google and Baidu charge clients for priority placement: pay, and you ranked higher.

Search optimization became a profession in its own right. SEO, search engine optimization, became a recognized specialty. By 2006, search engine marketing (SEM) in the U.S. was growing far faster than traditional advertising or any other online channel.

Google had become the internet's biggest entrance, and it moved to entrench that position. In 2009, Google merged the address bar and search box in its Chrome browser, which until then had been two separate fields. In a blog post, Google said users could simply type a keyword where they once typed a web address, press Enter, and see the results. The approach became standard: Microsoft added something similar in IE9 in 2011, where searches defaulted to Bing, and Safari 6.0 in 2012 also merged its search and address bars.

This was the inevitable end of making the internet more visual, simpler, and better suited to how people actually behaved. If you could reach any page from the results, why pay attention to the domain at the top? The decentralized, egalitarian ideal of the early Web was slipping further and further away.

Shopping, reviews… more "search engines" appeared, each reshaping how you find information in its own niche.

After Google, a string of companies with the same idea followed.

In 2004, the year Google went public, Yelp was founded as a search engine for restaurants. Two former PayPal employees, Jeremy Stoppelman and Russel Simmons, wanted to build a site to help people find local businesses online. As the founding story goes, one summer Stoppelman came down with the flu, searched online for a doctor, and turned up almost nothing useful. Yelp built on local services, joining social networking and online reviews so that users could ask others to recommend nearby businesses.

Yelp quickly became a city guide for San Francisco's tech workers, who in turn generated huge amounts of content for it. Yelp gave that content a template — a fixed structure, like a blog post. Before 2010, Yelp was adding more than ten cities a year, building a vast database of restaurants, hotels, and landmarks.

TripAdvisor in 2000, Dianping in 2003: by deepening the information in a single niche, these platforms displaced the official websites that businesses used to build for themselves. To find somewhere to eat in a given city, you searched Yelp, which also offered what an official site never would — most importantly, ratings and user reviews. The process was cleaner and more natural than opening a search engine, hunting for local information, and picking through a list of sites.

E-commerce is, at heart, also a place that aggregates information for people to search. In 2003, the year before Google's IPO, Amazon was selling $6 billion of goods a year online — a vast database of product information. By the same logic as search advertising, merchants wanted to pay Amazon for a higher ranking.

By 2013, Amazon's advertising revenue had reached $614 million. It was no match for Google's, yet Amazon had one direct advantage: it profited straight from its own e-commerce, held far more granular data, and tied purchasing behavior to ad targeting more tightly. Google's advertising depended on users searching for something and meeting a relevant ad; Amazon knew not only what shoppers browsed but how often they rebought a given category of product. None of that data was within Google's reach, and none of it could be turned into purchases by Google.

A similar bidding-based ad model was already established in e-commerce. Taobao, for instance, sold ad products such as cost-per-click (Through Train), cost-per-thousand-impressions (Diamond Booth), and cost-per-sale (Taobao Affiliate). Alibaba's advertising revenue last year already reached 108.4 billion yuan.

These were "search engines" within particular verticals. Whether a merchant page on Dianping or a store page on Taobao, each was still a web page at its core — just one whose domain nobody noticed anymore. Then came social networks, which made the web page itself optional and chipped further at the value of search.

In 2006, the two-year-old Facebook launched News Feed and Mini-Feed. In earlier versions, Facebook had only profile pages; people posted updates to record their lives and occasionally clicked over to a friend's page to see what they were up to.

At first News Feed drew complaints from privacy-minded users. But Facebook's user count doubled within a year. Over the next decade Facebook kept News Feed and kept refining its interest-recommendation algorithms — it owned the user data, served precisely targeted ads, and spared users the trouble of searching: the feed delivered information to them, and delivered exactly what they liked and wanted to see.

On News Feed's tenth anniversary, Mark Zuckerberg said: "News Feed is one of the biggest bets we've made over the past 10 years, and it's helped shape our community and almost the entire internet."

Chinese social networks moved quickly to follow the News Feed model. QQ Space, Kaixin001, Renren, and Weibo were all built on the feed.

Advertising changed too. It could now sit right in the feed, next to your friends' updates, with no website needed at all. You no longer needed a search engine to reach a web page. The unit of content was no longer a page but a post.

And the domain name? It had lost every reason to be seen or remembered.

The rise of apps sped up the monopoly. What you can see is decided by one or two companies in each domain, and more and more businesses don't even have a website.

After Google, Amazon, and Facebook had each cornered their markets, the smartphone arrived.

The smartphone era began with the iPhone in 2007. Starting from a device that was easier and more intuitive to use, it brought apps, services, and the mobile internet to the masses. Personal computers had been around for more than a decade, yet as of 2005 only 16 percent of the world could get online. By 2018, according to KPCB's internet report, half the world's population could — most of them on phones. The average adult smartphone owner spent 5.9 hours a day staring at the screen.

The spread of the internet only tightened the big companies' grip. Some of the niche "search engines" from the PC era were a natural fit for mobile — Dianping and Yelp, say, could surface nearby businesses based on your location wherever you were. That location data also gave rise to new kinds of advertising.

The big companies swelled fast over these ten years of mobile internet. In 2018, seven of the world's ten most valuable companies by market value were internet companies. The mobile era strengthened the firms in the business of advertising and traffic distribution: Facebook had more chances to gather data and track user habits, and so could sharpen its ads further. Mobile now accounts for nearly two-thirds of all internet advertising revenue. Advertisers spent $30.9 billion on mobile media in the first half of 2018, up 42 percent from $21.8 billion a year earlier.

According to Business Insider, Netflix topped the iOS App Store revenue rankings at $790 million a year. In fact, seven of the top ten apps by revenue were streaming content platforms, including Tencent Video, iQIYI, Kuaishou, and YouTube. App Annie's report reached a similar conclusion: the top ten apps by consumer spending were either entertainment or social.

It is hard to call these platforms web pages in any meaningful sense. Whether social networks or content platforms, they hold enormous power over what users actually see. Every niche has a dominant player controlling half the market: Google monopolizes search worldwide; Facebook and Tencent monopolize social networking globally and in China; Amazon and Alibaba dominate e-commerce globally and in China.

This no longer looks like the internet as it was first imagined. Today, for a business to reach its audience online, the answer is not to build an official site, optimize for search, and let people click links. Businesses have to put their products onto the giant aggregator platforms — Dianping, Taobao, JD.com, Amazon, Trip.com Group, Booking, Facebook, WeChat, Toutiao, and so on.

Online attention is concentrating on a handful of apps, nowhere more than in China. WeChat has more than one billion active users, roughly 15 million official accounts on its platform, and more than 5 billion followers in total — meaning one in every 40 Chinese people runs a WeChat official account. These accounts have nothing to do with web pages; they have no domain of their own outside WeChat. And, of course, making them disappear is entirely within WeChat's power.

WeChat is not the only one pulling things off the open web. Merchant information on Dianping no longer has its own domain; Taobao store URLs belong to Taobao; Toutiao allows links to third-party pages in principle but, in practice, recommends almost only content hosted on its own servers.

Nearly all online business has moved onto a centralized internet. What Trip.com Group, Meituan Dianping, and Taobao sell may be wildly different, but what they do is identical — they decide what information users can see, and they profit from it.

"If you'd asked me ten years ago, I would have said the internet is great, it connects everyone, but I was wrong," said the Web's creator, Tim Berners-Lee, at a summit this year. "Rather than going from website to website, now people are all on the same website… People don't need to jump around. What they see is decided by the people who write the algorithms for social networks."

Cover image from Unsplash @ Yiran Ding