Our primary principle is to work as a team to deliver the best outcomes for the business.
We combine face to face and verbal communcation with individual focus and use tools such as Slack, Trello, and GitHub to communicate openly within a project, within Axurcio, and publicly.
While we are considerate of people's feelings and opinions, we cannot let those concerns get in the way of getting the right outcome for the business and the success of our work. We'd rather be too honest than too polite.
Our standards are very high, and bringing on a new team member requires a "yes" from everyone who participated in the interview process. Therefore, we expect the best from each other, give each other the benefit of the doubt, and encourage each other to take initiative to improve ourselves to improve my contribution to the team and the company.
We recognize that we can always be better. Therefore, we have strong opinions, loosely held, and take initiative to improve ourselves, the company, and our community. We are proponents of the Agile manifesto
We work at a sustainable pace. We work four days for the business and one day on "investment time." We typically spend Monday-Thursday on business work and Friday on investment.
We believe in working wherever and whenever we work best. Because we rely on each other we confirm if we cannot meet a commitment, including letting our team members know if we cannot get to the office.
We know face to face communcation with our team is paramount & commit to spending at least 3 days in the office per week for conversation.
When taking time off during work, we discuss how it will impact the schedule with other team members.
We communicate when and where we work to our team through Slack.
Investment time is time for investment in ourselves, our company, and our community. Primarily this means doing something that interests us like building tools for our team, contributing to open source, researching & presenting on a topic, attending community events, or just improving the workplace zen. The goal is to encourage individuals to improve and share their knowledge with the rest of the team.
We organize our investment work on the "Team Time" Trello board.
One of our primary process goals is to make frequent, small releases of our working software.
This section describes how we achieve one week's worth of iteration on a product. It lays out the process we follow and some communication tactics.
Being a virtual team we do onlines 'standups'
We mention what we are doing. And how we can make things more awesome.
We do this in order to:
We try to have 4-6 hours in the work day that overlap for synchronous communication.
For some people, it is sometimes difficult to disengage from work when working at home. Also, flexible hours means that sometime one may work non-traditional hours in the day. We should be conscious to keep to a sustainable pace and take a break away from work.
What gets measured gets done
We measure ourselves using the following metrics
Here's the flow. Read our git protocol for the git commands.
Test-Driven Development moves code failure earlier in the development process. It's better to have a failing test on your development machine than in production. It also allows you to have tighter feedback cycles.
Code reviews that happen right before code goes into master offer similar benefits:
Martin Fowler has an extensive description of Continuous Integration. The basics are:
This helps ensure there's nothing specific to the developer's machine making the tests pass. The code in version control needs to run cleanly in production later so before the code is allowed to be deployed to production, it is run on a CI server or service.
When a build fails, we get alerts in Slack and via email. Click the alert and we see a backtrace that gives us a hint of how to "fix the build."
When we write the fix and commit to version control again, we'll get a "passing build" alert in Slack and via email. Click the alert and we see the passing build.
Green is good.
A solid test suite is an absolute requirement for a web application in our opinion. However, one major problem with test suites is that they get slow as they get large.
CI can ease the pain by distributing the test runs in parallel. We've had 45 minute test suites cut down to 2 minutes using this technique.
We've used CruiseControl, Team City, TFS and other CI platforms/libraries that we manage ourselves. This resulted in many hours of expensive attention.
We use AppVeyor
CI test runs are triggered by GitHub post-receive hooks. The hooks we have on most of our GitHub repos are:
We live in a magical modern era where many problems have already been solved for us. We focus on the core product as much as possible and outsource operations as much as possible to external services.
This saves time and money. We can get started using those services in minutes, and pay a service tens or hundreds of dollars per month instead of paying developers thousands or tens of thousands.