Update: We now have an active Request for Proposal for an Independent Researcher to tackle interesting questions pertaining to developer productivity & satisfaction. If you know a researcher who would be a good fit, please mention our posting to them? We'd be especially eager to work with professors/academics who operate well at high speed.
Assuming happy developers are productive developers, shouldn't there be more research by now on what makes them happy?
Yes, there are good developer studies already out there. We eagerly consume the best whenever we come across them. Which isn't often. We figure that since GitClear is the company writing the most about "developer productivity" and its relation to "developer happiness," we should be contributing to the public corpus of knowledge on this subject. Even if a manager or developer isn't a GitClear customer, there's going to new conclusions about how dev teams operate at peak efficiency, peak satisfaction.
If you'd like prefer to skip the rest of this article and jump straight in as a volunteer, thank you! Please fill out the < 5 minute survey we've created here. We will get in touch with you as soon as our experiments have been designed. You'll get one of a limited number of memberships we're giving away to help support this research. It might take a month or two to get the Researcher and experiment selected/designed, so your patience is much appreciated. 😄👍
In informing how we'll configure our developer survey, we'll be looking to some of our favorite existing developer studies from 2020. If you're interested in this sort of thing, we'd highly recommend checking these out:
Stackoverflow 2020 Developers Survey. 65,000 developer respondents makes this the gold standard of developer research, Stackoverflow deserves the 🏆 for contributing all this knowledge back to the public domain. Insights we found especially Interesting or usable: most beloved programming languages; writing that first line of code; most used frameworks; most wanted web framework; most dreaded platform; baseline developer job satisfaction; only 25% of developers not interested in new job opportunities; job hunt factors; and salary by developer type.
JetBrains: The State of Developer Ecosystem 2020. Survey conducted by a worldwide leader in thoughtful IDE development. Especially useful or interesting: programming languages by usage; devs jump at the opportunity to participate in open source; average product team 2-7 people; and 37% of companies practice some form of pair programming.
Hired.com State of Software Engineers 2020. Most concerned with developer hiring and interests, in accordance with its source. Interesting and useful: 66% believe code exams irrelevant to the job; and slight preference toward remote over daily office life.
These surveys are great for exploring what languages developers are using, but they don't reveal much about the inner workings of what makes a contented or productive developer. That's where we hope to assist. Here are some questions we're setting out to answer:
How does a developer's self-reported happiness with their job correlate with the amount of legacy code in their repo?
Goal: Determine whether tech debt has a measurable impact on developer satisfaction with their job. Assess correlative behavior patterns a manager might utilize to spot if their team suffers from this condition.
How does increased use of Slack & Microsoft Teams impact team productivity and code quality?
Goal: Illustrate how the most productive teams wield their instant communication powers.
What are the measurable habits of new Developers who are judged by their managers to have been a "great hire"?
What does the documentation and code recency of the project look like in those cases? Is there advice that can be given to new hires on how to make a positive first impression?
To what extent does adding developers to a team slow down team velocity?
Goal: What happens when new developers are tossed onto an overdue project? Conventional wisdom says that tacking developers on an overdue project slows it down, but managers keep doing it, so let's understand when it works. This question might also help startups make more informed decisions about when to scale up their dev team.
How does pair programming effect code velocity and quality in a repo?
Goal: Help guide manager intuition around when programming in groups is a net positive for a team's code quality & velocity.
And also some other questions.
To what extent does low Line Impact velocity correspond a Senior Developer's evaluation of where tech debt in a project resides?
Goal: Arm Senior Devs with evidence that (in the form of measurable data) to persuade management a problem exists that needs to be remedied.
To what extent does Line Impact correlate with Story Points?
Goal: If there is high correlation, then we can combine these two so as to offer help when work on a ticket is behind, without the developer having to ask.
You get the idea. This is about half the ideas we have prepped for our first round of research.
Since we know it can take valuable time from a company to participate in studies like these, we feel it prudent to offer an inducement beyond the 6 month membership.
Our pledge is that we will publish the results of all the studies we conclude, whether they show what we'd consider “significant results” or not. There has been a lot of deserved scrutiny on how publishing bias limits the useful of research, and we want to take all possible steps to avoid contributing to that. We will encourage our Independent Researcher to pre-register the hypotheses to be tested, since that seems to be today's gold standard in how valid, useful results can be replicated.
If you're interested in these questions, or if you're just hoping to see more transparent experimental data on what makes developer teams function optimally, we would love to have your help.
To start, we ask that you please take 5 minutes to complete our participation screening. If you lead developers, or if you are a developer led by a data-minded team leader, the survey is designed to go fast.
We will start emailing selected participants their 6-month membership code within the next month or two, once our Independent Researcher has been selected and designed their first round of experiments. If you know any professors or academics that are passionate about the field, please ask them to drop us a line at hello@gitclear.com with the subject "Research Position." We will work collaboratively to figure out an efficient & transparent process by which to collect and distribute new data.