Company goals around using, consuming, contributing to and maintaining open source software (OSS) projects are usually related to market positioning, required upstream features development, and talent attraction or retention. Well-organized OSS efforts are managed by an Open Source Programs Office (OSPO).
To achieve these goals and respond to increasingly specific requests from the OSS community and ecosystems, OSPO teams need to organize and combine a number of data sources, analyze information faster, and deliver more insightful analysis from several channels. This post explains the benefits of having a single source of truth and some ideas OSPOs can apply to achieve it.
One of the issues of having data about open source projects in different places is that they cannot be correlated or analyzed together. This can lead to incomplete or inconsistent views of the OSS ecosystem and community and poor decision making. For example, OSPO teams might end up being pulled in different directions and insights.
Common mistake: The difference between single source of truth and single version of truth
Single source of truth and single version of truth are often used interchangeably, but the two are not synonymous. In order to better understand the importance of having a single source of truth within OSPO teams, let’s first clarify these terms.
A single source of truth is “the practice of structuring information models and associated data schema such that every data element is mastered in only one place” (Wikipedia). The key here is that all relevant data is available in only one place. In contrast, analyzing software development process data by using different platforms with different ways to gather and visualize results (e.g GitHub stats, Discourse stats, Docker stats, Slack stats, etc) refers to multiple sources of truth.
A single version of the truth is “the view that everyone in the organisation agrees that data from one singular provider is the real and trusted figure” (Wikipedia). That is, for instance, having one software development analytics provider within OSPOs, rather than many. In contrast, a lack of a single version of truth could result in debates about which numbers are correct and whether they can be trusted.
We assume that an OSPO team can agree on the metrics and we assume that a single version of the truth is given. This post focuses on the benefits and disadvantages of having a single source of truth for an OSPO.
Strategy: Prerequisite to effectively using a single source of truth
We need to highlight that, even though OSPO teams agree on having a single source of truth, if they don’t have a clarity of purpose first, they will drown in data without really knowing how to communicate with it.
OSPO should first agree to use the same, unified source for all their data needs across the entire organization (data integrity). This will keep everyone aligned on high-level purpose and focus on the concrete results expected from each OSPO area.
A good example of this is having a homogeneous and documented metrics definition for each OSPO team including key performance indicators (KPIs) and objectives and key results (OKRs). For instance, relying on a consolidated community that establishes meaningful metrics within OSPOs, the same way CHAOSS does for open source health.
Once the strategic vision is clear, it is then time to drive the attention on the different data sources needed to get the most accurate picture possible. Open source ecosystems involve continuous interactions between different communities, happening in more than one platform. Identifying which platforms are those and be able to untangle key relationships, becomes crucial to improve ROI.
However, depending on how OSPO is structured within a company (legal requirements, departments involved etc.), having a single source of truth might not be the best solution. Here are some pros and cons:
Pros of having a single source of truth
Teams might spend too much of their time copying data from one system to another. Every time somebody manually inputs data is another opportunity for errors to be introduced. This might cause work expenses and burn-out consequences. A single source of truth automates this process and collects all relevant data in one place.
OSPO teams that suffer from data fragmentation often find that their ability to innovate and implement new systems is slowed. A single source of truth allows uncovering relationships and insights that may lead to innovations otherwise not obvious.
Breaking down silos
When different teams work in silos, it is easy to forget to share insights and data. A single source of truth can break down siloed working as all departments have access to the same insights.
Managers have an informational role to inform others in the organization. A single source of truth offers one consistent view of the same metrics across the organization and reduces the chance of leaders inflating their teams’ results.
Create a cohesive view
When tracking data across different platforms, it is harder to identify trends and understand the whole picture of what the data shows. A single source of truth provides a complete and cohesive view of a community or ecosystem.
Cons of having a single source of truth
No limited access
Using one single source of truth means that all team members who use this platform will see all the available data, even though some of the data may not be needed by all of the OSPO team members.
More data needs to be gathered and enriched to be displayed in a single platform, which demands more time to process, which leads to slower performance of the metrics platform.
Creating a single source of truth built in-house from scratch might take a lot of time and effort from any OSPO team. Having a service provider that has a solution ready and maintains the platform to also fix future issues and build custom integrations reduce costs and of course, time effort.
Bonus point: Single source of truth with one or multiple dashboards?
We have seen there are many benefits of choosing a single source of truth for your OPSO data-strategy, but once we cover that, there might be companies that prefer having multiple dashboards for specific needs while others see more convenient using just one single dashboard.
Here is a small table to see some differences of having one vs multiple dashboards
|One Dashboard||Multiple Dashboards|
|Pro: Cross-project analysis||Contra: No cross-project analysis|
|Pro: Maintain and update only one dashboard||Contra: Changes to visualizations have to be replicated in each dashboard|
|Contra: Slower performance because more data||Pro: Faster performance because fewer data|
|Contra: A user of the dashboard can see all data||Pro: Users can be given access only to some of the dashboards.|
At Bitergia, we have faced both situations among customers: Some community building specialists prefer a single dashboard for its DevRel strategy while others prefer multiple dashboards.
Single source of truth resources for OSPO teams
Open Source Program Offices (OSPO) use Bitergia Analytics as a single source of truth and single version of the truth about their ecosystem of open source projects. Bitergia Analytics helps managers in their decision making and reporting by analyzing open source communities and processes. Learn more about how Bitergia helps OSPOs through their metrics journey.
If you want to get deeper knowledge about OSPO metrics, check out our free webinar: OSPO metrics journey for open source sustainability.