Enhanced Messaging Workshop at AAAI

I'm in Chicago for the AAAI 2008 Conference's Enhanced Messaging Workshop. Greg and I will be talking about some of the driving ideas behind Xobni, the hardships of building commercial email software. We'll also demo some products from our skunkworks.

I'm hoping that this workshop will hopefully fill an important gap: There are some conferences about spam fighting on one end, and general machine learning / data mining / user interfaces on the other end, but no forum for academic discussions of fighting email overload. See you there!

Startup School: Surfing the Wave

The most interesting pair of talks at Startup School were Greg McAdoo from Sequoia, followed by David Heinemeier Hansson from 37 Signals. McAdoo talked about how to build a billion dollar company, and Hansson spoke about how to create a small profitable one.



The contrast couldn’t have been bigger: The Sequoia talk was about big words and surfing metaphors. The ridicule in the video's comments shows that the audience was displeased. Hansson, on the other hand, wanted small web companies that charge users. A simple model. People loved it.

In reality, the two talks were about different things. Hansson talked about how to create a profitable business with a web site. This is a well-known if underused model. McAdoo, on the other hand, was talking about building the next Cisco, PayPal, Yahoo, or Google. Billion dollar businesses, not million dollar ones.

In order to make a really successful billion-dollar startup, you do have to have find a large, growing technology market. Marc Andreessen agrees. Xobni, for example, found great market when we changed our focus from an analytics package to today’s sidebar, a product people use for hours a day. We’re now in a much better, billion-dollar market, have a wider audience, and greater odds of success.

So why the disconnect? Two reasons.

First, I think that smart technical people – the kind that inhabit the computer science halls of Stanford, Berkeley, MIT, CMU, ETH and Oxford – have a strong appreciation for beautiful technology: The beautiful technology that, for example, made Google what it is today. But it’s not the technology that created that success, but the fact that people wanted fast full-text web search, which was just happened to be very hard to build. The market did it, not technology.

Second, when business guys mention "identifying customer needs", "designing a platform", and "assembling great team DNA", hackers shrug. Why use these big terms for talking to customers, drawing up architecture diagrams, and recruiting that great hacker? Because these are abstractions. Just like you wouldn't discuss each for-loop when describing what that method does, VCs use these abstractions to communicate effectively. It often takes experience to really understand the meaning behind them. Once entrepreneur, you’ll understand what it means to identify markets and customer needs, and you won’t need the baby steps spelled out for you. You’ll be using those big words yourself.

Xobni at CEAS

We'll be down in Mountain View for the Conference on Email and Anti-Spam on Thursday and Friday.

Want to chat about Xobni? Watch out for us in the black Xobni shirts! Or drop me a line if you want to meet up.

BlogCampSwitzerland

When Corsin and I started organizing BarCampZurich last year, we had lofty ambitions: I remember sitting in an organizers’ meeting in a basement in Chur, Switzerland, explaining to a room of skeptics that we’ll have 50 or more attendees. It worked out pretty well, though: Almost 100 people showed up. (Here's a summary of how BarCampZurich worked out.)

When I heard that Dominik Tarolli et al. are setting up BlogCampSwitzerland - an unconference for European bloggers - I was on the side of the skeptics: "They’ll have 50 bloggers, maybe," I thought. So far, they have more than 150 signups!

I hope the European blogger crowd will be as interesting as the roomful of geeks who met last October. I’ll try to be there and may even host a discussion.

BlogCampSwitzerland, March 24, 2007, 10 am - 4:30 pm, ETH Zurich.