I’ve learned about computers from my dad. He brought home as TRS-80 Model I when I was about 8 or 9 years old, and we had a series of those early computers in the house from that time onwards.
For some reason BASIC didn’t catch with me, but I did a lot of programming in dBASE and acquired an early appreciation for high-level languages and database architecture. I started my career as a paid programmer at the age of ten, making $10/hour coding mailing list management systems for local businesses.
In college I got my first exposure to the new Macintosh computers and learned how to make maps with an early version of Freehand and it’s closer relative, Pagemaker. I did a lot of freelance work as a Mac expert in those days. Font Manager, hell yes! I then returned to San Francisco to help my father launch a company using Uniface, one of the major high level database-driven development environments of the day. Although that company didn’t get far off the ground, it was my launching point for a very fun couple of years doing development, software architecture, and management consulting at Compuware. I don’t know how I knew what I was talking about, but it seems that I did, because our projects were successful and clients liked us.
Before long, however, I got tired of the cross-country flights and, since I was dead set against an automobile commute, I decided to find a cool software job that I could walk to from my where I lived in North Beach. This was in 1996 – there were no tech companies in San Francisco. Still, I stuck to my criterion of walkability, and after several interviews I landed a job at HotWIRED the “digital” web arm of WIRED magazine. HotWIRED wasn’t a digital version of the magazine, it was a company designed from the ground up to publish original digital media unique to the web. My years at HotWIRED (later WIRED Digital, Lycos, TerraLycos) taught me a ton about the web and about building web applications, but also about media and media businesses.
It was during the end of my time at HotWIRED that I started AdMonsters. I had gotten heavily involved with the tech infrastructure that served the ads that HotWIRED was trying to make money from, and I went to a couple of the early conferences that covered that field. And they sucked. Oh man did they suck. They were literally soul-sucking. Luckily, I wasn’t the only person who thought so, and when I called around and proposed that a few of us get together on our own for something that didn’t suck, AdMonsters was born. Read on in the Conferences section for more on that story.
Back to software. When I started AdMonsters I had already turned down at least two jobs that were solidly on the CTO track, and my tech chops suffered at the expense of learning new things, like accounting, sales, marketing, conference production, and everything else that comes with running a business. But you know what’s awesome? When you run your own business you get to do everything. So I got to be CTO too, which mostly meant that I got really good at Drupal, and retreated into the guts of our server whenever the other aspects of the business got too frustrating.
My background in technology served me well during the AdMonsters years. Unlike print advertising, online advertising is a highly technical way for a business to make money, and understanding how technology works from the ground up and the top down is key to understanding the business models of online publishers. As we later came to say, OPS is how online media works.
I was never much of a fan of advertising, and these days I’m known by some for saying that “advertising is obsolete”. I don’t have much to do with the huge and insanely complex world of “ad tech”, but I am still fascinated by what we can do with computers and with software.
I agree that “software is eating the world”, but also that the best technical solution is often no technology at all. These days almost everything we do is enabled, controlled and connected by software, and yet most of us have not idea how complex and how downright messy these systems are. Messy means fragile, expensive, prone to failure, slow to change and hard to fix. And yet: anything is possible with software, and we are just at the beginning of the evolution of computing. Siri is not an AI, but we will soon have real AI’s, and they will have a lot to teach us.