Why Ubuntu succeeds: Shuttleworth isn't an uber-geek
Open source tends to be developers writing for other developers. When will we break that cycle?
I really liked this InformationWeek article with Mark Shuttleworth, founder of Ubuntu. I liked it in part because I think Mark is an exceptional person. But I also liked it for its insight into why Ubuntu has done so well.
Hint: It has a lot to do with making a very geeky thing (Linux desktop) less geeky.
IW: Why are you concerned about ease of use when many Linux developers see it as a lesser issue?
Shuttleworth: I'm not brilliant technically like Linux Torvalds. I feel the gaps in Linux usability personally. When you're an exceptional developer, you can be blinded to the ease of use issues. Besides, people gravitate to the problems that they're best equipped to help fix.
That said, Mark is brilliant and quite technical. We had dinner together last week in London, and Mark was trying to convince me that I could write code if I wanted to. No, I really can't/couldn't. I find that developers tend to underestimate just how isolated and special their talents are. They seem to think that we all share that talent but choose to deploy it elsewhere.
I don't write code because I can't. I wish I were a Luis Villa with an understanding of code and business issues, but I'm not. I'm lucky to get the latter - the former is a complete mystery to me.
Which is why open-source development communities really need to find ways to bring non-techies into the development conversation. We can offer all sorts of insight into how we use software, even though we're worthless at advising on how to code the software.
Back when I was in school, Microsoft used to come by to watch me work. They send all sorts of people into the field to study actual use of computers and software to determine how to make their products better. One can argue whether they do a good job of implementing that intelligence, but the data-gathering is excellent.
The open-source community can learn from that. We should be much more in tune with the needs of the lay user of technology. That is, in fact, the premise underlying open-source software. Users of software developing for other users of software. But open source, when not tied into a corporate-sponsored project, tends to fail the average user.
We can do better. Mark points the way as to how. Get more non-technical users involved. Any ideas as to how?