Cross-communities collaboration: A new era to develop communities

Since Open Source has been declared a key factor for gaining competitive advantage within Software Industry, more and more companies begin to develop and collaborate within Open Source projects. This implies an important growth within project ecosystems, with more complex interactions between projects. Thus, are community professionals prepared to face these changes that have a direct impact on how we develop communities?

According to IDC:

By 2022, over half of the code in 75% of new apps will be from external sources. The need to improve developer productivity is driving the use of open source and commercial code repositories

This means 75% of the code will be done by third parties (which includes Open Source). Being able to understand what’s going on with projects out of your company seems crucial for gaining competitive advantage now that such a big amount of code relies not only within the walls of a single organization.

The inter-dependency of Open Source projects has changed the nature of Open Source community development. As communities converge, release schedules and priorities collides, project leaders need to adjust OS models, re-think interactions with multiple release cycles and juggle the divergent agendas. Thus, new strategies raise up in order to overcome these changes

For instance, we can focus on specific community roles, such as the community manager. This role has been initially in charge of one single project, taking care of it’s community health. However, with this new era, the community manager needs to face more complex environments where Open Source has created more interdependent relationships, which means the need of a more systematic data-driven approach to ease community development.

Community Management
The community manager role needs to face more complex environments where Open Source has created more interdependent relationships

New entities: Open Source Program Offices

As more companies contribute to Open Source, there is an increasing amount of community roles, some of them more focused on growing user community such as Developer Relations or Brand Evangelists and other ones focused on growing developer and contribution community, such as Community Managers. With Open Source being so important now within companies, where can these type of roles fit within an organization?

Some companies are creating Open Source Program Offices (OSPO) in order to gather all these roles and to have a centralized space where manage their open source plan and strategies and help them to align Open source activities with company goals.

Moving from Art to Science

As mentioned before, convergence of communities means managing a wide amount of data. “Art” practices are no longer enough when making decisions. Things like personal interaction, perceptions, opinions, and gut feeling lose power while “scientific” ways such as observation, systematic approaches, hypothesis or statistics that base decisions on facts work better to manage big amount of information effectively.

Companies are investing a vast amount of resources in open source, and they want to maximize the return of their investments (ROI) and they need help to better identify dynamics driving new technology initiatives. We help in all of this process to those people responsible for the community development efforts, by applying systematic data-driven approaches to community development to help identify inter-dependencies between projects, people, and entities. By applying the insights they get with Bitergia Analytics, they can adapt their community models to the new realities and complexities of the new era of corporate-sponsored open source cross communities collaborations.

Analyzing Inter-dependency

During the whole post, we’ve been explaining why there is an inter-dependency among Open Source projects, and how this is changing the way project ecosystems interact with communities.

A project is not about just one single group of people contributing, but a complex network full of connectors. Analyzing and getting insights from this network can be used as a powerful tool for many Community Managers or DevRel Professionals. For instance, we can better identify our “Personas” or in this case, the different group of of people with similar behaviors and characteristics that are in a community.

Let’s see this with an example. In the next image, we can see the different connectors between projects (Kubernetes, Jaeger, Open Tracing, ContainerD, and OpenShift) from three different companies (Red Hat, Uber, and Rackspace). These connectors are people who have knowledge and show interest for more than just one project. We can zoom in and even recognize the identity of these connectors, and the amount of activity they are doing.

Cross-Community Management
Chart details: Dots are contributors, blue rectangles are projects, dot size is the amount of repositories and the edge thick is the amount of commits

The next visualization shows the main connectors within Uber between Jaeger and Open Tracing.

Network analysis: Uber and Jaeger and Open Tracing
We can identify four connectors between Open Tracing and Jaeger within Uber and see its total activity

As you can see, with this data-driven approach, we can dig from the whole network ecosystem (the big picture of cross-communities network) to the details. The next image shows the whole ecosystem from CNCF and OpenShift.

CNCF and OpenShift
CNCF + OpenShift Ecosystem

To Sum Up

There are some important points we can learn from this:

  • No company is working in just one “thing”
    They need to be aware of interactions and collaboration among communities. For this reason, upstream coordination is essential.
  • Relationships matter
    Even though Open Source Communities need to move towards a more data-driven approach to take better decisions. We shouldn’t ignore personal interactions, that still provide us us a very important qualitative information.
  • Community Management vs Community Development
    Because of the big amount of information and data we need to process, Community professionals might start thinking about communities development instead of trying to manage them.
  • Data matters!
    It’s crucial to have a data strategy for collaborative projects development to gather facts and insights if you wanna make better decisions.

At Bitergia, we specialize in software development analytics, and this “convergence of communities” topic is one of things we are working on with Diane Mueller, from Red Hat. There is an introductory video to this topic from last DevConf in Brno available:

We are working with Red Hat on this, there was an update during last Open Source Leadership Summit (PDF slides available), and there will be more during next months. So, if you are interested, please let us know and/or leave comments to this post…

Leave a Reply

Powered by WordPress.com.

Up ↑

%d bloggers like this: