Before writing iStumbler I worked at various startups during the dot-com bubble in San Francisco. After the 'big pop' of 2001 I suddenly had a lot of free time and started writing iStumbler to develop my programming skills. Currently, I'm a stay-at-home dad and work on iStumbler and other projects when there's a minute to spare.
iStumbler is the leading wireless discovery tool for Mac OS X. What makes it special?
Besides the wireless network detection iStumbler also allow you to monitor your Bluetooth devices and Bonjour Services in one simple interface. Some features, such as the ability to browse wide-area Bonjour domains and the recently released Spectrum Widget, don't appear in any other application.
How many people use iStumbler? How many contribute to the project?
The web site averages well over 20,000 unique visitors a month, and I'm seeing roughly 30,000 downloads of each version. There are two kinds of contributors: Subscribers who pay for early access to new versions of iStumbler, and code contributors. In all, nearly 600 people have pitched in to help support the project as subscribers and we've had two code contributors, one of which is currently active.
A few months back you noted that iStumbler may stop being free. What led to that decision? Did this situation change your view on developing open source software?
This has been a tough decision to make, there is a feeling on the Web that when you offer something for free it should stay free. I'm not sure that's entirely fair to content creators, but the sentiment definatly exists and factored into my final decision: the current feature set of iStumbler including AirPort, Bluetooth, Bonjour and the Location plugins will always be free and Open Source but future plugins will be part of an 'iStumbler Pro' package.
Publishing iStumbler has completely changed my view of Open Source development. While I still believe in Open Source there is a lot of work to be done in finding sustainable business models for smaller projects. The service model promoted by ESR really only works for enterprise open source and popular consumer packages like Linux distributions, it's not a tenable option for a small ISV which markets directly to consumers. In fact, if you write a really useable piece of software, the service model works against you because users don't need assistance!
What is the most interesting fact you've become aware of while developing iStumbler?
Watching the wireless industry has brought two things to my attention: UWB and Mesh. Much more than Wi-Fi these two technologies in combination provide a credible threat to the existing, centralized, telecom infrastructure. UWB brings massive bandwidth and power efficiency which, when applied to a Mesh Topology, could create a multi-gigabit wireless distribution network that will not require the expensive central offices and billing systems used to support existing phone and data carriers.
What's your take on wardrivers? Some say they're harmless while other label them as criminals.