Write Code Every Day


Last fall, work on my coding side projects came to a head: I wasn’t making adequate progress and I couldn’t find a way to get more done without sacrificing my ability to do effective work at Khan Academy.

There were a few major problems with how I was working on my side projects. I was primarily working on them during the weekends and sometimes in the evenings during the week. This is a strategy that does not work well for me, as it turns out. I was burdened with an incredible amount of stress to try and complete as much high quality work as possible during the weekend (and if I was unable to it felt like a failure). This was a problem as there’s no guarantee that every weekend will be free – nor that I’ll want to program all day for two days (removing any chance of relaxation or doing anything fun).

There’s also the issue that a week between working on some code is a long time, it’s very easy to forget what you were working on or what you left off on (even if you keep notes). Not to mention if you miss a weekend you end up with a two week gap as a result. That massive multi-week context switch can be deadly (I’ve had many side projects die due to attention starvation like that).

Inspired by the incredible work that Jennifer Dewalt completed last year, in which she taught herself programming by building 180 web sites in 180 days, I felt compelled to try a similar tactic: working on my side projects every single day.

Illustration by Steven Resig

I decided to set a couple rules for myself:

  1. I must write code every day. I can write docs, or blog posts, or other things but it must be in addition to the code that I write.
  2. It must be useful code. No tweaking indentation, no code re-formatting, and if at all possible no refactoring. (All these things are permitted, but not as the exclusive work of the day.)
  3. All code must be written before midnight.
  4. The code must be Open Source and up on Github.

Some of these rules were arbitrary. The code doesn’t technically need to be written before midnight of the day of but I wanted to avoid staying up too late writing sloppy code. Neither does the code have to be Open Source or up on Github. This just forced me to be more mindful of the code that I was writing (thinking about reusability and deciding to create modules earlier in the process).

Thus far I’ve been very successful, I’m nearing 20 weeks of consecutive work. I wanted to write about it as it’s completely changed how I code and has had a substantial impact upon my life and psyche.

With this in mind a number of interesting things happened as a result of this change in habit:

Minimum viable code. I was forced to write code for no less than 30 minutes a day. (It’s really hard to write meaningful code in less time, especially after remembering where you left off the day before.) Some week days I work a little bit more (usually no more than an hour) and on weekends I’m sometimes able to work a full day.

Code as habit. It’s important to note that that I don’t particularly care about the outward perception of the above Github chart. I think that’s the most important take away from this experiment: this is about a change that you’re making in your life for yourself not a change that you’re making to satisfy someone else’s perception of your work. The same goes for any form of dieting or exercise: if you don’t care about improving yourself then you’ll never actually succeed.

Battling anxiety. Prior to starting this experiment I would frequently feel a high level of anxiety over not having completed “enough” work or made “enough” progress (both of which are relatively unquantifiable as my side projects had no specific deadlines). I realized that the feeling of making progress is just as important as making actual progress. This was an eye-opener. Once I started to make consistent progress every day the anxiety started to melt away. I felt at peace with the amount of work that I was getting done and I no longer had the over-bearing desire to frantically get any work done.

Weekends. Getting work done on weekends use to be absolutely critical towards making forward momentum (as they were, typically, the only time in which I got significant side project coding done). That’s not so much the case now – and that’s a good thing. Building up a weeks-worth of expectations about what I should accomplish during the weekend only ended up leaving me disappointed. I was rarely able to complete all the work that I wanted and it forced me to reject other weekend activities that I enjoyed (eating dim sum, visiting museums, going to the park, spending time with my partner, etc.) in favor of getting more work done. I strongly feel that while side projects are really important they should not be to the exclusion of life in general.

Background processing. An interesting side effect of writing side project code every day is that your current task is frequently running in the back of your mind. Thus when I go for a walk, or take a shower, or any of the other non-brain-using activities I participate in, I’m thinking about what I’m going to be coding later and finding a good way to solve that problem. This did not happen when I was working on the code once a week, or every other week. Instead that time was consumed thinking about some other task or, usually, replaced with anxiety over not getting any side project work done.

Context switch. There’s always going to be a context switch cost when resuming work on a side project. Unfortunately it’s extremely hard to resume thinking about a project after an entire week of working on another task. Daily work has been quite helpful in this regard as the time period between work is much shorter, making it easier to remember what I was working on.

Work balance. One of the most important aspects of this change was in simply learning how to better balance work/life/side project. Knowing that I was going to have to work on the project every single day I had to get better at balancing my time. If I was scheduled to go out in the evening, and not get back until late, then I would need to work on my side project early in the day, before starting my main Khan Academy work. Additionally if I hadn’t finished my work yet, and I was out late, then I’d hurry back home to finish it up (instead of missing a day). I should note that I’ve been finding that I have less time to spend on hobbies (such as woodblock printing) but that’s a reasonable tradeoff that I’ll need to live with.

Outward perception. This has all had the added benefit of communicating this new habit externally. My partner understands that I have to finish this work every day, and thus activities sometimes have to be scheduled around it. It’s of considerable comfort to be able to say “Yes, we can go out/watch a movie/etc. but I have to get my coding in later” and have that be understood and taken into consideration.

How much code was written? I have a hard time believing how much code I’ve written over the past few months. I created a couple new web sites, re-wrote some frameworks, and created a ton of new node modules. I’ve written so much I sometimes forget the things I’ve made – work from even a few weeks prior seem like a distant memory. I’m extremely pleased with the amount of work that I’ve gotten done.

I consider this change in habit to be a massive success and hope to continue it for as long as I can. In the meantime I’ll do all that I can to recommend this tactic to others who wish to get substantial side project work done. Let me know if this technique does, or doesn’t, work for you – I’m very interested in hearing additional anecdotes!

Discuss this post on Hacker News.

Posted: April 10th, 2014


If you particularly enjoy my work, I appreciate donations given with Gittip.

73 Comments (Show Comments)



Comments are closed.
Comments are automatically turned off two weeks after the original post. If you have a question concerning the content of this post, please feel free to contact me.


Secrets of the JavaScript Ninja

Secrets of the JS Ninja

Secret techniques of top JavaScript programmers. Published by Manning.

Ukiyo-e Database and Search

Ukiyo-e.org

Japanese woodblock print database and search engine.


John Resig Twitter Updates

@jeresig

Infrequent, short, updates and links.