The Gossip, Recklessness and Freedom of 20 Years Ago Made the China of Today | Memoirs of 1999, Part One

We interviewed several of the earliest chroniclers of the Chinese internet, hoping to reconstruct that era through them.

No one could have guessed that the shape of Chinese business today would be set almost entirely by companies founded 20 years ago.

Sohu, founded August 1996.

NetEase, founded June 1997.

Tencent, founded November 11, 1998.

Sina, founded November 30, 1998.

Alibaba, founded April 1999.

Trip.com Group, founded May 1999.

Dangdang, founded November 1999.

Shanda, founded November 1999

...

On a partial map of the internet startups of 1999, we can find these companies that still carry weight today:

NetEase was the most-visited site of the time, its email and personal homepage services booming. Ding Lei, then 28, rushed from Guangzhou to Beijing to collect an award; four years later he would sit at the top of the Hurun Rich List.

Zhang Chaoyang graduated from MIT and returned to Beijing to become the most famous internet entrepreneur of the age. He renamed his company "Sohu" — a name that echoed, in spirit, the internet legend "Yahoo."

Wang Zhidong was named president and CEO of Sina. His biggest achievement until then had been building a Chinese-language platform for Windows and founding Stone Rich Sight, a software company. Now he was turning to internet business — less than a year before Sina would list on Nasdaq.

Tencent launched the instant-messaging software OICQ, later renamed QQ. It was not the only product on the market with a name like that; all of them imitated ICQ, made by an Israeli company. OICQ was laggy at first. Two years later, in an office in Huaqiangbei, Shenzhen, Pony Ma was struggling to raise money.

Jack Ma came back to Hangzhou from Beijing, and although what Alibaba actually did was still unclear, he made the cover of Forbes the next year. What people remembered about him most was that his English was good.

Trip.com Group was founded in Shanghai; among its co-founders was Neil Shen, a master of capital operations.

Dangdang copied Amazon and began selling books online. The next year, Kingsoft, whose general manager was Lei Jun, turned Joyo.com's main business into an online bookstore as well.

That same November, Shanda Networks was founded. Two years later, the Korean online game Legend of Mir 2, which Shanda held the exclusive license for, would redefine both popularity and addiction.

Behind these companies stood a whole crowd of peers and rivals who have since vanished and been forgotten. Some had their moment and then went quiet; others rewrote the rules of the game, only to be overturned by the next wave of innovators...

We interviewed several of the earliest chroniclers of the Chinese internet, hoping to reconstruct that era through them.

Some of them compare it to Woodstock; some remember a freedom and an absence of limits they had never known before; some praise it for opening a whole new way to build wealth.

"Courage" and "persistence" are the words that come up most often — more often even than "ideals." And, of course, "luck."

Others say that what looks unbelievable now felt entirely ordinary at the time; the people who would later become startup heroes and models of wealth were just the ordinary people around you, and back then you could not have picked them out of a crowd.

We will publish these chroniclers' oral accounts and their own stories one after another. This is the first.

In 1994, Hong Bo was in his early thirties, a documentary director at the August First Film Studio. Everyone knew he loved literature; soon he would win a literary magazine's top short-fiction prize with his story Fishing Rod (《鱼竿儿》), and not long after that he would lose all interest in becoming a writer.

That April, China was fully connected to the internet. That same year, Hong Bo went to Zhongguancun for the first time, walked past long rows of stalls that looked like a wholesale market — motherboards, cases, hard drives, CPUs, memory sticks — and assembled his first personal computer out of the parts. Building it this way cost two or three thousand yuan less than a brand-name machine, but it still took nearly all of Hong Bo's savings.

Dialing up over a phone line, Hong Bo logged onto CFido, a dial-up bulletin board system, and became one of its well-known core users. The sysops of several CFido nodes were named Pony Ma, Ding Lei and Lei Jun — men who would go on to become towering figures in the internet industry. When one fellow board member got married, Hong Bo and Lei Jun jointly gave him a washboard, wrapped in fancy paper so it looked like a high-end keyboard.

Two years later, Hong Bo clicked open his first website, becoming one of the two or three hundred thousand internet users in China at the time. The internet replaced literature, giving him a pleasure he could hardly describe. "This is wonderful," he decided — this was "something worth giving a lifetime to."

yahoo.com — that was the first website Hong Bo ever opened. People knew little about websites then; Yahoo was the exception. Originally called "Jerry's Guide to the World Wide Web," the search site was created by Jerry Yang, an ethnic Chinese student at Stanford. Its click count passed 10 million by the end of 1994, and its market value shot up after it went public in 1996. "Jerry Yang, the Chinese entrepreneurial hero, was the hottest internet myth of the time."

In China, though, no internet wealth story had been born yet. The two tallest buildings in Zhongguancun still belonged to Lenovo and Stone Rich Sight. Lenovo made most of its money distributing well-known foreign computer brands; Stone Rich Sight had started building Chinese-language websites and forums — the earliest form of what would become Sina — but people then thought of it as a software company, and saw the new website and forum business mainly as a way to sell more of its RichWin Chinese-language platform software.

For people like Hong Bo, everything was just beginning. "You got online, and what opened up was an American website. That feeling was extraordinary." A few years earlier there had been almost no way to reach information from outside China; one of the few was tuning a specially modified radio to "enemy stations" like Voice of America — propaganda-tinged, and still compelling. The internet changed that: instead of receiving propaganda, you went and got the information yourself. Starting from Yahoo, you never knew where you would end up an hour later. There was no jamming, only complete freedom and flow — even if the bandwidth was narrow and the connection slow.

The China Internet Network Information Center, CNNIC, was founded in 1997. As the only statistics anyone could cite at the time, CNNIC reported that in those first years the number of internet users doubled every six months. There were 620,000 users in October 1997 and 1.175 million by July 1998. Infrastructure held the growth back; otherwise it would have been faster. Getting online meant dialing a telephone, and telephones were not yet common in homes. Because the telecom carrier charged private ISPs (internet service providers) steep agency fees, the cost of getting online never came down. Chinese-language websites had thin content, too — there were thousands of them by count, but few that users visited regularly.

In 1998, Hong Bo took on another, better-known name: keso. On his personal website China Download, he offered software downloads and ran a column called "keso's Ramblings." "Keso" sounds like the "cough" alert in online chat rooms.

keso wrote his views on software and the internet industry and gave users advice. He stood for a kind of person the internet had produced — "not a professional," not an entrepreneur in any real sense, but "an observer of the industry, observing things on everyone's behalf." New software and new websites appeared every day; people were not yet curious about who made them, but they wanted to know how to find, use and judge them. In a 1999 poll, China Download was named one of China's ten best websites.

keso wrote about whatever he liked. In 1999 he warned people about the Blaster worm; his advice amounted to little more than don't download software carelessly, don't visit strange websites, and patch Windows promptly. A few years later, as internet companies grew and "new economy" and "knowledge economy" became fashionable words in business and government, keso mocked 263 Mail on the platform DoNews for starting to charge and driving its free users away. Another piece that made a mark, "I Don't Know That Man" (《我不认识那个人》), attacked Chen Tianqiao for flying a nationalist flag to steer opinion in a commercial dispute: "He sits on the stage. Sits surrounded by spotlights and flashbulbs. I cannot make out his face, or the memorial archway behind him. I have to leave as fast as I can, because I feel sick. As I back away, I keep telling myself: I don't know that man."

In the two years between, the internet went through a roller-coaster rise and fall. Entrepreneurs had never drawn this much attention. Sina, Sohu and NetEase finally joined the Nasdaq startup-legend club in 2000, but almost at the same moment their share prices and the Nasdaq index both dived. Between March 2000 and October 2002 the entire US IT industry lost 5 trillion dollars in market value. Yahoo's stock fell from $118.75 in January 2000 to $8.11 in September 2001. Half of these companies, it later turned out, survived past 2004, but at the time the media raced to announce the end of the world. In China, Burning.com (《烧.com》), published in 2001, declared that this would be "China's biggest economic bubble" of the entire 21st century. In hindsight, of course, the claim was wide of the mark.

keso never published a piece questioning the internet itself, and he rarely came down hard on the early internet entrepreneurs. He misses those first years of the internet in China, at least somewhat — surprised to find that he himself held "an extreme passion" for it, and that "when so many entrepreneurs bet their whole lives on it, that makes you feel it's a good place to be." Looking back at China's early internet startups, keso compares them to Woodstock, and to Burning Man.

Two weeks ago we met keso in Beijing. He was wearing his signature red-star cap, the same one Cui Jian wears. He owns seven or eight of them. Today he writes about the internet on his own WeChat official account.

To help you understand him better, here is his own account:

I started using the internet in 1996. Getting online meant dialing up over the phone, which tied up a phone line, so first you had to apply for a telephone. Applying was expensive — I remember that installing a landline in Beijing then cost 5,000 yuan (RMB) in installation fees alone, and going online was billed at local-call rates. My salary was decent, around 5,000 yuan. Very few people in 1996 made 5,000 yuan a month, and at my peak I spent as much as 4,000 yuan a month on internet access alone. You can imagine what an expense that was.

That year a friend and I took part in remaking a magazine. IDG had invested in a Tianjin magazine called Software (《软件》). We all know IDG today as a venture capital firm, but back then it was a company that invested in computer media — that was its whole business. It had also invested in China Computerworld (《计算机世界》) and Popular Software (《大众软件》).

Software was founded in the early 1970s and originally just published papers. After IDG invested, it wanted to turn the magazine into a stylish tech title. They had seen how hot Wired (《连线》) was in Silicon Valley at the time, and wanted to follow its model to popularize the culture around computers.

We did not stop to think how hard this would be — if it could be done, we figured, we'd give it a try. We printed the magazine in full color on coated paper; apart from the fashion magazines, Software was the first full-color computer magazine. Every issue came with a CD of video material covering an extremely wide range of subjects. For one issue we got a huge batch of photographs from the photographer Tang Shizeng, scanned them and put them on the disc — he had shot a lot in Iran during the Iran-Iraq War. There was an overseas Chinese property developer who scanned the works in the Louvre in 3D and then printed them out; in the scan you could see a brushstroke here, a raised bump there. We filmed him too. None of it was directly about computers, but it stood for a fairly modern outlook, a new culture. The magazine also had no interest in speaking for business leaders.

Of course, after a year we found we could not keep it going. There were too few readers. We made the internet the centerpiece, but very few people had actually touched it. We closed the last issue at the end of June 1997 and left in July.

Six months later I joined NetChina, a private ISP. In the US, AOL was the largest ISP, and most Americans got online through it. Many of China's private ISPs were aiming to be "China's AOL." That included Yinghaiwei, the earliest private internet startup.

Yinghaiwei was pushing the idea of the internet to the public very early, including the billboard it put up on Baishiqiao Road, by the south gate of Zhongguancun: "How far are the Chinese people from the information superhighway? 1,500 meters north." At the 1,500-meter mark there was a "Yinghaiwei Time-Space" storefront where users could try getting online. I never tried it, since I could already get online on my own, but I saw the little shopfront. In 1997 I went to Shenzhen, and since I could no longer dial up to Beijing from there, I needed a local ISP, so I bought a Yinghaiwei account.

The private ISP founders of that time all felt the internet was a genuinely unbound, brand-new field. In every other field there were old powers with a lot of influence. If you wanted to go into textiles, say, the textile industry was all big state-owned enterprises, and breaking in would be hard. The internet had none of that — nothing at all — and the regulators did not understand it either. It was a very free space, and that was exactly what felt so good about it. It had no rules, no constraints, no regulatory limits; anything you could think of, you could do, you could experiment. People used to say the central government had carved out one special place, Shenzhen, where you could break through certain policy and legal limits. The internet went further than that — it had almost no limits at all.

But these ISP companies were rarely mentioned again, because the whole direction of the ISP business was wrong. China's ISPs were in effect agents for the telecom carrier: the carrier wholesaled bandwidth to you, and you sold users their connection to the internet — but the carrier was doing the same thing itself, at a far lower cost than yours. The bandwidth it wholesaled might run to more than a million yuan a month for two megabits (private ISPs later negotiated that down to a bit over 400,000 yuan a month). A hundred megabits costs almost nothing today, but back then two megabits might mean 2,000 people sharing the same pipe. If those 2,000 people did not bring in more than 400,000 yuan, you could not even cover what you owed the carrier for bandwidth. For the carrier, that cost was zero — the bandwidth was its own. What happened in practice was that the private ISPs had no room left to cut prices, because access fees had to keep covering the bandwidth bill. The carrier could keep cutting forever. So the private ISPs all died.

For many years in the early period, the name mentioned most was Zhang Chaoyang. For years he was the internet's number-one spokesman. Only later, once BAT and Shanda's Chen Tianqiao arrived, did he gradually recede. Of that first generation of internet people, he was the one who had studied in the US — Wang Zhidong and Ding Lei never studied abroad. Sohu was also one of the earliest of the three big portals. He was China's earliest evangelist for the internet, popularizing ideas like venture capital and giving people a fuller understanding of it. On top of that, his personal style was flamboyant and outgoing, and he liked meeting reporters. Ding Lei probably had few chances to meet reporters — reporters had never even seen him — but everyone had seen Zhang Chaoyang. The media needed a figure like this, and Zhang Chaoyang liked playing the part, so he stepped forward.

It is not that only Zhang Chaoyang could stand for the internet. Plenty of people could have spoken for it; many of them simply did not want to step out. Take Pony Ma: for years nobody saw him in the media, and even a decade or more later, many people had no idea who the man behind QQ was. Some people, on the other hand, are naturally high-profile. Back when nobody knew what Alibaba did, Jack Ma was already holding his "West Lake Sword Debate" in Hangzhou, and the guests he invited were exactly the people who were the biggest names on the internet at the time.

Set against the traditional IT entrepreneurs, this generation is clearly different. The earlier group mostly came out of the state system. China had only just opened up its economy, and some entrepreneurs with courage and vision were the first to feel the opportunity, so they came out. But the first half of their lives was spent inside the system. Liu Chuanzhi, for instance, was at the Institute of Computing Technology under the Chinese Academy of Sciences. Peking University Founder Group is a Peking University enterprise, and Tsinghua Unigroup is a Tsinghua enterprise.

The entrepreneurs who came later mostly had nothing to do with the state system. Lei Jun, for example, joined Kingsoft straight out of university, and Kingsoft was never a state-system company — it was private from the first day. Whether you went through that state-system stage or not matters a great deal. The traditional IT entrepreneurs had far more constraints to break through, because you did not know what was allowed and what was not, what you could do and what you could not; you had to feel your way. The ones who came after had it much easier.

But by the time the internet arrived, new problems may have appeared.

Take the government's restrictions on internet investment: how do you get around the limits on foreign capital? How do you dodge them? You had to keep so-called domestic internet companies free of foreign interference while still taking money from abroad. China carried a lot of leftover ideological baggage here — whether something was "capitalist" or "socialist"; whether a foreign shareholding, if it got too high, touched on national sovereignty. A lot of these ideas seem strange today, but that is how people thought then. The internet industry started out the same way. Even now, especially when you take foreign investment, the problem is basically still there: you have to design a VIE structure, in which the domestic operating entity is a purely domestic company that holds the various government licenses and has no foreign capital in it, but that passes its profits to the foreign company through a set of agreements. Everyone had to build a structure like this, purely to get around this law.

Sina ran into exactly this when it went public. There were many policy minefields, and someone had to walk through them to find out which spots had no mines and which paths were closed. Someone had to do that work, and it ate up a lot of time. Sina's listing dragged on so long that it was not settled until Nasdaq was already close to collapse. By the time Sohu and NetEase listed, the collapse was under way, and their IPOs were dubbed "bloody listings." Some companies were held up completely — 8848, the e-commerce company, never made it, never listed, and eventually died. It died for many reasons, of course, but failing to get approval was part of it. Internet companies really did have to break through all kinds of visible and invisible limits.

In the early years, some people in the government thought the internet was a good thing and others thought it was a threat; opinion inside was split. In fact, the internet taking root in China was a fairly accidental affair. That is why many people say someone like Hu Qiheng, who pushed for civilian use of the internet in China, deserves enormous credit (note: Hu Qiheng was then vice president of the Chinese Academy of Sciences; in April 1994, at a meeting of the China-US Joint Committee on Science and Technology Cooperation, she restated China's request to the US National Science Foundation to connect to the internet, and it was approved). It is because people like her persuaded a section of government officials to agree to open this door. The government did not understand the internet well at first either; some people in the industry told officials what the internet could be good for. The government found that appealing, and hoped a knowledge economy would drive the whole national economy.

Seeing real benefits took many more years. The harms showed up quickly. The information controls that had worked in the past stopped working; you needed a new set of tools to handle a national gate that had suddenly swung open. Before, there was no technology for blocking foreign websites, and nobody had even thought of it — but it appeared soon after, and the system only grew tighter and stronger.

So later the government stopped worrying. If you can control something, you do not need to worry about it. In the early 1990s, when it had only just started, they were somewhat worried, because it was an unknown. I have never dealt with these officials directly, and I have no interest in doing so.

After the internet bubble burst in 2001, I felt a lot of media outlets were a bit gleeful about it: "Look, you blew such a big bubble and now it's popped." "I always thought what you people were doing was pretty unreliable anyway." In fact, that bubble looks like nothing by today's standards. Look at Nasdaq now — it is above 7,500, with a peak around 7,900 — whereas back then, at this same point, it was only just over 4,000.

It looks as if the bubble burst and left nothing behind. That is not true. Yes, the bubble burst and a lot of money burned, but it left plenty behind. First, the users stayed. The users the "boom" had drawn in stayed, and they kept using the internet. A lot of infrastructure stayed too, like the dark fiber Google bought at the time (dark fiber being fiber-optic cable that was never switched on). Once people decided the dark fiber was useless, Google bought it cheap and later used it in its broadband business. Infrastructure like that can keep paying off for years, even decades. Some technology stayed, and the people who had been trained stayed.

It was also impossible for the bubble not to burst. People say that back then you did not even have to do anything — just rename your company something-dot-com and you could raise a pile of money. That was obviously not normal.

The media likes to say "the collapse of internet speculation," but I think the speculation was mainly the investors': the investors saying the internet is hot, I want to invest in the internet. It had little to do with the entrepreneurs.

There was not much speculation among the entrepreneurs themselves at the time. Like Woodstock, or Burning Man today, there was not much commercial intent in it; it was more a kind of self-indulgence. It did not need to carry some commercial payload. What it carried was what everyone wanted in their ideals.

For one thing, they were different from the traditional IT companies. Take Lenovo — Lenovo is an interesting company. To my mind, Lenovo is the representative company of Zhongguancun, but it is not a technology company. Lenovo was mainly a trading company; at the time it was the biggest distributor for AST and Compaq. Later it spun the distribution business off and built PCs itself. But the PC industry is not even counted as high tech in the US any more. It is manufacturing.

In April 2000 Lenovo built a website called fm365. It never became anything, but Lenovo did make money from it, because building fm365 let it raise a lot of money on the Hong Kong stock market. Investors bought the internet concept wholesale: Richard Li and Li Ka-shing built tom.com, which had nothing behind it, and still, bang bang bang, pulled in huge sums on the market. If the internet concept was this hot, why not use it to raise money?

The young entrepreneurs were different. They had no other path. Doing the internet just meant doing the internet, not using the internet to raise a pile of money for something else. Relatively speaking, it was purer.

Some people's thinking was very simple: build a piece of software and sell it for money. Ding Lei, for instance, figured he could make money selling email software, and the software was quite simple — he did not need to do much development, mostly localizing a foreign open-source program, making some changes to suit the local market, and then he could sell it for anywhere from a few hundred thousand to more than a million yuan a copy. That money looks like nothing today, but back then even a "10,000-yuan household" counted as rich.

Take Pony Ma. He built a piece of software and wanted to sell it to Runxun for pagers. QQ was called OICQ, the "online pager," back then; he said messages could be sent online straight to a pager. Runxun and tom.com both said they did not need it, so he and his team had no choice but to keep going on their own and look for a buyer.

Apart from selling out, nobody could see how this industry could make money. People had a lot of fear about starting a company back then. So the ones who actually threw themselves in and dared to start a company had extraordinary courage; it was not a small thing. You could not see where the money was. You just knew that reading the news on Sina was great and using NetEase's email was great — but how did these companies stay alive? Nobody knew. They did not know themselves.

So when the bubble burst in 2000 and outside investment suddenly stopped, they had to start making their own money, and that was when they discovered they did not know how to make money at all. Some of them listed successfully, which did show that the internet could produce enormous personal wealth — there is no doubt about that, and it drew later entrants in, wave after wave. But while listing looked like a glorious moment for that first generation of internet companies, for years afterwards they mostly looked battered. The moment they listed, the bubble burst, and what followed was simply working out how to keep themselves alive.

Luckily, China Mobile launched Monternet around then. Monternet saved them. Ringback tones, phone wallpapers, customized text messages... NetEase launched "Same-City Dating," which let you arrange dates with other people in your city by SMS. A lot of non-internet businesses rescued the internet companies.

Compared with the internet entrepreneurs who came after, that first generation of companies also had a very different temperament. People said at the time that the internet was like a pastoral age — the internet's idyll. The internet entrepreneurs who came later were looking for the next hot thing and grabbing the opportunity, rather than doing what they liked or wanted to do. I much prefer the original kind, the person who rushes in before there is any way back, over the person who only comes in once they can see there are plenty of opportunities and money to be raised.

For the first kind, what matters most is believing that what you are doing has value. Pony Ma believed from the start that this thing had value; when you can see your user numbers rising every day and usage climbing, you cannot think it has no value. Most people feel that without a way to raise money and a way to make money, they cannot believe in something's value. Plenty of people were building ICQ clones at the time — Sina and Sohu both tried it, NetEase too — and none of them stuck with it; they gave up on their own. Only Tencent kept going, kept going while nobody believed in it, and kept going until everybody did believe in it, by which point it was too late for anyone else to get in.

Hong Bo

Also known as keso. Got his first personal computer in 1994 and started using the internet in 1996. In 1998 he launched the column "keso's Ramblings" on his personal site "China Download," writing his views on software and the internet industry. His DoNews blog "Playing the Lute to a Cow" was widely influential; he continues his internet commentary today on the WeChat account "What keso Thinks." He took part in the relaunch of Software magazine and once worked at NetChina and the IT site ChinaByte.

Cover image from Markus Spiske, Markus Spiske on Unsplash,