Why I left a great job to build a crappy task list

“I find it hard to believe you are leaving just to build a crappy task list”

I spent nearly all of the previous decade taming the chaos of a successful online dating system, namely PlentyofFish. I joined a very small tech team in 2011 and together we learned the hard way how to turn a startup into a mature software development shop. In those early days we all wore multiple hats and spent much of our time putting out fires and averting disaster. In the years to follow we expanded (and incessantly reorganised) the team and introduced (and frequently circumvented) new process. Scaling the software was trivial, scaling the operation, not so easy.

In the latter half of the decade, PlentyofFish was acquired by the Match Group and I accepted two new challenges. The first was to join the executive team with a grand new Chief Architect title, and the second was to begin work on tearing apart the monolithic system we wrestled with each day, and attempting to rebuild it as a set of event-driven subsystems (some might say “microservices”). The original technical debt investment which had given us a hugely successful product, a great office and rewarding jobs finally had to be repaid. I suggested that the project would have a "half-life" of 18 months. Four years and many grey hairs (on my face of course) later, the first few buds of success were flowering. All that work was finally beginning to pay off. That’s when I made the decision to leave.

During the years at PlentyofFish prior to the acquisition, I had a little side project: a simple productivity tool called Redo (like "todo", but repeated) that would help me manage the less exciting parts of work and life that were all too easily neglected. Built in AngularJS with a Go back-end hosted on Google’s App Engine, it also gave me a chance to learn a few new skills and to flex my somewhat neglected programming muscles.

The aim of this tool was to capture the things you want to stay on top of and then motivate you to get them done. Each item has a rough check-in cadence as opposed to a specific date and time. These then appear on a filterable list that is ordered and colour coded to show you the most pressing items first. The soothing goal being a sea of green coloured items on your list meaning you are up to date. Gradually, the list reorders and things that are becoming due bubble to the top and change to more urgent hues. You mark each item complete, optionally with a note, or you snooze it for later. The notes act as a journal so the next time you catch up with a specific work colleague, prepared for the weekly meetup, or checked the status on a project, you can be reminded where you previously left things. A few minutes once or twice a day reviewing your list and marking a few complete, et voila, a sense of calm and fulfilment descends upon you.

An example Redo List

A handful of late night hacking sessions and Redo reached a point of being good enough to be useful. From then I diligently dog-fooded it every day for a few months, mostly to check-in with important members of the team at PlentyofFish. Eventually I stopped using it. Maybe it was the lack of push notifications or maybe it was just how overwhelming the list could look when you failed to update it for a while. Perhaps it was just a bit too ugly and lacking those dopamine inducing micro-rewards that made Candy Crush so successful. I wasn’t sure.

Ever since I read Scattered Minds by Gabor Mate I knew I was a textbook case of Attention Deficit Hyperactivity Disorder ADHD and this helps explain why I thrive when the context keeps switching but struggle when the chaos is tamed. I believe I’m an effective "wartime" software developer (please excuse the mutilation of Ben Horowitz’s Peacetime CEO / Wartime CEO insights), I don’t delight in disasters, but I do delight in averting or resolving them. I find it significantly more challenging to complete mundane tasks such as filing expense reports or any sort of form filling in fact, to the degree that I would regularly prefer to be out of pocket than work out what number needs to go in what box. I also have a diabolical memory which can be challenging when engaging in the gentle art of management if you are unable to recall a crucial detail from a previous conversation with one of your reports.

I’ve also struggled with anxiety for as long as I can remember. Half of all adults with ADHD also have some sort of anxiety disorder. It’s really not an awesome combination as the anxiety can nag you to do things that the ADHD won’t allow you to do.

Redo was a meagre attempt at a tool that can help me at work while easing my symptoms of ADHD and anxiety and I’ve long wanted to resume the project, find out why it didn’t work and research what might work better. Over the last 12 months I’ve been learning a great deal about mood disorders, anxiety disorders and neurological disorders, and it's interesting how much of an overlap between the symptoms there are of many disorders and diseases. Perhaps by focusing on the most common symptoms, I could build something useful that also helped people cope a little better with the stresses of 21st century life?

In his book Feeling Good, David D. Burns gives a great intro into Cognitive Therapy (here's a nice Ted Talk of his if you're interested). It's a fascinating approach to rationalising thought processes to combat and reverse crippling negative thoughts. Would any of the techniques described in the book work as part of a mobile application? Perhaps, but before I dust off my programming hat, I'm going to need to learn a great deal more about Cognitive Therapy and (confusingly) the group of therapies known as Cognitive Behavioural Therapy (CBT). I'm also going to have to talk to other human beings who know significantly more than me about these things.

So I'm going back to University in June to learn more about Cognitive Behavioural Therapy. It's an online course at Wilfrid Laurier University in Toronto. Meanwhile I'm doing a separate Cognitive Behavioural Therapy Practitioner course on the online educational platform Udemy.

I don't know where this will lead. Maybe I'll find a few extra features for my "crappy task list" or maybe I'll build something completely different. I may discover there is a team out there already doing good work like this that I can join, or maybe I'll hang up my programming hat for good and become a therapist. We'll see.