Net Homeless

I first went online in 1997. It was a completely different place. No social networks, no smartphones, no corporations controlling most of the traffic. The internet was more like a vast, unstructured network of enthusiasts.

Since the late 1990s, I have watched almost every stage of its development. FidoNet, IRC, the first forums, personal home pages, early search engines, the arrival of blogs, the birth of SEO, the rise of contextual advertising and social networks, and the gradual monopolisation of the web that followed.

In the early 2000s, like many people at the time, I began experimenting with ways to make money online. It was an era when “startup founder” was not yet a familiar label, and people working from home over the internet were often seen as layabouts or dreamers.

There was a whole movement of so-called “internet hobos”: people who publicly described their attempts to build a small online business and make a living entirely on the web. I took part in that movement too.

I mainly created and developed networks of websites. These included satellite sites for selling links, informational projects funded by contextual advertising, and experiments with different ways of monetising traffic.

There were many link exchanges at the time: Sape, MainLink, TrustLink, and others. The link market was a huge industry. You could create dozens or even hundreds of small sites, gradually improve their metrics and rankings, and then monetise them by selling links.

At the same time, I built sites for Google AdSense. Contextual advertising was only beginning to gain momentum, and many niches were almost empty. Find an interesting subject, write good content, attract search traffic, and a site could start bringing in real money.

Much of the work in those years was a constant series of experiments. Building my own content management systems, automating publication, and writing parsers, content generators, and tools for analysing search positions and link profiles.

I had to do a great deal myself because ready-made solutions either did not exist or were extremely primitive. The internet in those years was much freer in general. A small project made by one person could compete with large companies.

One successful site could grow to tens of thousands of daily visitors in a few months. One good automation algorithm could replace the work of an entire team. Many of the technologies and approaches I work with today, including automation, large-scale data analysis, complex distributed systems, and artificial intelligence, have their roots in those early experiments.

My entire subsequent career grew out of simple curiosity about how the internet, search engines, and large information systems work. I experienced the internet almost from the beginning of its mass adoption and watched it evolve for more than two decades, from a small community of enthusiasts into one of modern civilisation's most important infrastructures.