Dimagi Engineering spends a lot of time on Github, and with that, comes lots of comments and emojis. After one year at Dimagi, I’ve become an avid emoji user and relish the opportunity to use emojis in the most astute way possible. There’s just something great about being able to express your thoughts in a small set of pictures. At Dimagi we use them quite frequently and in increasingly creative ways. This is a deep dive into how our developers rocked out with emojis in 2015.

We racked up a grand total of 1601 emojis in the dimagi organization. That’s 4.38 emojis a day and 0.0000507 emojis a second. Some people like to use emojis more than others though. Who on our team is the most avid emoji user?

Screen Shot 2016-02-24 at 11.10.11 PM

2015 has been a tumultuous year, and emoji usage was definitely not uniform. In fact, I have reason to believe that emoji usage spurs more emoji usage until it peaks at some maximum output of emojis per user. Below I show how emojis usage has changed by month:

Screen Shot 2016-02-24 at 11.10.36 PM

What does the break down look like for each user?

Screen Shot 2016-02-24 at 11.11.52 PM

Indeed we can see a self reenforcing emoji pattern. At least with the avid emoji users, the more you use emojis, the more I use emojis. There are some outliers. Ethan (esoergel), a heavyweight when it comes to emoji usage, goes through highs and lows throughout the year.

Sadly, the truth behind increased emoji usage may have other causes that are less poetic. Our lint-review bot used to give a thumbs up emoji every time a pull request was updated and had no lint errors. Midway through the year we decided to drop that functionality and just add a label when the pull request had no lint errors. This meant that nothing would alert the reviewer when the pull request was updated. To counteract that, our team developed a habit of writing a comment that contained a :white_check_mark: emoji to alert the user. Or, you can believe that emojis spur more emojis like I do.

However, emoji usage is more nuanced than the simple binary indicator: do you use emojis or do you not use emojis. I’m interested in how Dimagi uses emojis. What are the emojis that Dimagi most uses?

Screen Shot 2016-02-24 at 11.10.52 PM

The above shows the most frequent emojis with the exception of the :white_check_mark: because it dwarfs all other usages. We’ve used that emoji 555 times this year. It’s how we indicate to the reviewer that the tests have passed for the pull request and is ready for review and/or merge. I’ve also removed emojis that were used less than 7 times to improve readability of the graph.

Everyone has their own style of writing, and so too with emojis. What user uses which type of emojis?

emoji_distinct

It’s interesting to look at the diversity of vocabulary used when employing emojis. Here we can see that Cory and I share a passion for pushing the boundaries of emoji icons. Yet, in absolute emoji usage, Cory blows me away.

Crafting an emoji sentence is one of the more rewarding ways to use emojis. Who on our team uses the most emojis in a single comment? Our top 5 emoji comments use 24, 17, 9, 8, and 8. Here’s what they look like:

Screen Shot 2016-01-14 at 4.58.14 PM

Screen Shot 2016-01-14 at 4.59.12 PM

Screen Shot 2016-01-14 at 5.00.53 PM

Screen Shot 2016-01-14 at 5.01.48 PM

Screen Shot 2016-01-14 at 5.00.40 PM

Biyeun certainly captures the trophy for the longest string of emojis. However let’s examine a few emoji sentences that capture a deeper meaning, which cause the reader to read between the lines. Here are some of my favorites:

The good ol’ “throwing you under the bus” emoji. Sorry Tyler.

Screen Shot 2016-01-14 at 5.13.23 PM


Abstract art

Screen Shot 2016-01-14 at 5.13.44 PM


Our website is getting faster (it really is)!

Screen Shot 2016-01-14 at 5.14.55 PM


Lock away those cookies and make our site safer!

Screen Shot 2016-01-14 at 5.15.07 PM


If there’s anything I’ve learned about emoji usage over the past year at Dimagi, it’s not the size, shape, length, or even the type of the emoji; it’s the love that comes with it.

Screen Shot 2016-01-14 at 5.24.58 PM

If this inspires you to analyze your own team’s emoji usage, I’d like to help you get started. This data comes from GitHub’s data archive on Google BigQuery. Check out this blog post on how to use it. And if you want to get all of your emojis you can take a look at my query. Note this does not include the desciption text of a pull request. Good luck and have fun!