The Path to the Glukit 1.0 Preview

People who know me can attest to this (I hope): once I commit myself to doing something, I do it. Even if it makes my life miserable. Even if no one is looking and the cost of giving up is nil.

I do it because I need to create something. I do it because I know that the result might be worth it. I do it because I know that working on something brings a great satisfaction only on successful completion of it.

I started working on glukit in March 2013. We’re in August 2014, 16 months later. After an initial functional app, I had a good idea of what the minimal set of features I wanted in the 1.0 preview release. And I wanted to get that first release out. So I committed myself to making 1 commit every day until I got all of it done.

This got me a streak of 200 days of commits on github. The funny thing is that I realized only after getting hooked that my streak would only ever be visible to me because my repositories are private. But that’s not why the feature is powerful. There might be a social aspect to it but, to me, this was another behavioral science trick to help me get to a desired outcome.

The Dark Side

Now what? I’ve gone through the making of successful creative projects like this one before. The hard part if always the life after. The truth is that I feel exhausted from having worked so hard without a pause for so long. That hard work and dedication got me to a significant milestone. But, now at that milestone, it’s hard to find the energy to commit to the next phase.


The first day of my 2 weeks family vacation was the day I made the commit that marked the 200th day. I knew I had to wrap-up before losing focus. I’m afraid that taking a break will get me to move away enough that I won’t want to get back to it. I might get drawn to something new like drawing or writing a book.


But glukit is also different. I did commit before I left that I would return from vacation to announce the preview release of glukit. If I was successful in creating a convincing presentation of my idea for a better, behavioral science powered diabetes application, maybe you’ll have something to tell me. Comments, suggestions, requests. I worked on glukit for so long that I can’t see clearly anymore if it’s any good or if it can be effective like I imagined it.

You’ll tell me.

If you think it might have potential, I’d be excited to hear it.

Your emails might be what fuels my next commit streak.