effects of going open source [08/11/2007 13:00:06]

Well, I got over my fear of going open source, and the code is now freely available on the web. You can browse the code for the cornerhost control panel and billing system over at the cornerops trac.

What's astounding to me is that I actually had someone volunteer to set Trac up for me. One of my customers, Chris Braybrook did pretty much all the work, and even set up a second trac for my web libraries [the workshop trac]. (Thanks a million, Chris!) He's even gone through and set up his own working copy of everything.

Once the code was up, I was able to point directly to code on the web. I've been wanting to set up an OpenID server for my customers, but been frustrated by the poor example code on the various openID servers out there. So I posted to Innercircle (the private list for cornerhost customers) and was able to point to the code showing how my control panel does logins, and ask for help integrating that with an OpenID server.

Another customer, Bradley Peters, rose to the challenge within 24 hours, and showed me exactly where and how to modify the example server so that users could log in with their cornerhost usernames and passwords. Turns out he didn't get much use out of my source code to solve the problem, but it was a lot easier to ask the question once I could point to the source code: I'd written a draft of the post a couple weeks ago, but decided not to send it because I couldn't show anyone what I was talking about.

It's only been a week, but I'm already seeing tangible benefits.

But what about a desktop app?

I have another program I've been putting a lot of time into lately. It's a desktop app for improving productivity. I've been dying to talk about it, but kept quiet because I was considering launching it a commercial application and I wanted to think through the marketing plan first.

Part of me still wants to make this thing a closed source app. I'm seeing dollar signs and hearing cash registers.

But: I can also see how keeping it open and free would mean a much larger user base, and putting me in the ideal position to market a niche service specifically to those users. (Plus I'd then just be adding to my portfolio of subscription web services, rather than trying to be both a service provider and a software vendor at the same time.)

I don't think I would have considered this second possibility for the desktop app if I hadn't opened up cornerhost. It may be that the biggest effect of going open source will be in changing the way I think about business.