When to give up your Open Source Project?

After recently reading Jeff Atwood’s plea to maintainers to not be douche bags. I have come to the realization that I am a horrible Software Parent. I have been writing open source code since 2000, and some has even enjoyed moderate success. But where I have been dropping the ball for 10 years is in responsibly letting them go.

Exhibit A:

My first project, IPM was just a simple multi-user todo list I needed for the first company I started. I put it up on sourceforge in 2000 hoping that someone way better than me at PHP would help me fix it. But instead I got my first dose of internet popularity. People actually liked it, and they wanted new features. NO! THEY NEEDED NEW FEATURES! It was all very exciting, and the project benefited from contributions to the code and documentation. It was great for everyone. Fast forward about a year. IPM is now being used by some groups at CERN, which makes my nerd gland swell, however I could care less about adding another task display feature. I had just stopped caring about it.

I would get emails from concerned users asking when I was going to fix a bug, or release a new version, and I found myself just ignoring the emails. Not maliciously, but just out of apathy and a deluded thought that I would get to them eventually. It turns out that wasn’t very healthy, and the project sits there on several websites now, neglected and sad. Untouched for almost a decade.

Exhibit B:

My next project to have any success was a PERL monstrosity meant to help out with the administration of a game server than I ran. The server was for Return to Castle Wolfenstein. It was a great server with about 30 people hanging out at any given time on any night. Of course being the internet, some people were fond of showing up and causing trouble. The native commands for banning someone or kicking them off the server were clumsy and unreliable. So I wrote a simple interface to the server that would read what I typed in chat and execute the commands directly via a socket to the server. It worked very well and eventually ended up on several hundred other servers. Eventually I created a statistics affiliate network out of that code base, and as a player community we shared our abilities with a thompson submachine gun with every player that had been on the network. This pre-dated stuff like Valve’s stats by about 7 years.

But then… I stopped caring about that too. I ported the bot to Call of Duty, but quickly lost interest there as well. And it still sits, waiting like a good dog for me to come home and love it again. But it’s just not going to happen.

Exhibit R:

I skipped a whole bunch of other projects over the last decade that you don’t care about. This brings us to my current project: Dwarf Therapist. I hacked on it for most of the end of last year, and it has come a long way. But the telltale signs of not caring are showing their ugly heads.

But this time, I’m still engaged enough to care about its fate. There are other contributors who would willingly take it over. I must decide how and when to do it. And what role I want in the future. I know I can’t bear to see another of my projects just slowly decay. How do other open-source authors handle this stage of a project?

Advertisement