Open Source Governance and the Rise of a New Open Health Movement

Hadrian ZbarceaIn my previous article, we covered issues related to open source licensing and the impact of those licenses on adoption and distribution. In this article, we will explore the models for growing an open source project, and the individuals/organizations that do this.

Governance

Aside from licensing, the two next most important elements to consider in open source are collaboration and governance. Decisions about these aspects have to be made pretty early and have long-lasting consequences.

There are three main models of governance. Project governance provided by:

  • A benevolent dictator,
  • A company (or a commercial entity or organization)
  • A non-profit organization.

Many small open source projects are managed/governed by a 'benevolent dictator' entity, who decides what contributions (patches) to accept or not, who gets what rights, and so forth. For small projects, this is usually not a bad model as it leads to increased development velocity. If the "benevolent dictator" is also wise, the project has good chances to survive.

For large projects, a more formal governance is required to stay viable, as the lifetime of the projects extend past the fleeting interests of an individual.

Commercial open source has found its growing niche, and for many companies, it just makes sense. One reason is that using open source is a deterrent for the competition because anything released to the open space is rendered useless as a competitive advantage while still allowing opportunities for upselling one's own product.

Another reason is that collaboration with other companies leads to cost reductions. In areas that are critical but not directly related to the core competencies of the business. This is ideal because the cost reductions are achieved without compromising market position or profitability. This model works well for general purpose projects, such as infrastructure of user interface related. Google, Netflix, and Facebook are notorious for open sourcing high quality projects. However it is hard to see this kind of governance for healthcare space in a successful and sustainable way because there is no entity in the healthcare space for which healthcare is not a core business activity.

Specialized non-profit organizations provide the most known and most viable way of producing and governing open source projects. There are many such organizations, from the well known like the Apache Software Foundation (ASF), the Linux Foundation, or the Eclipse Foundation, to small ones, like WorldVistA, OpenEMR, OpenMRS, DHIS2, LibreHealth (under the Software Freedom Conservancy), or OSEHRA, which are exclusive to the health space.

That said, choosing a governance model is not a simple matter either. Some foundations, like the Apache Software Foundation or OpenEMR, are run by volunteers who try to act in the best interest of the project. They may be employed by a commercial organization interested in said project (and, therefore, may have conflicting interests at times). Other foundations, like the Linux Foundation or OSEHRA, have other organizations (commercial or not) as members, and decisions are usually influenced by the commercial interests of the members.

While I believe the volunteer run organizations have an edge, the other foundations are not far behind, as most of time, the commercial interests are well-aligned with those of the open source projects. At the end of the day, the passion, skills and motivation of the open source software engineers is what counts (which is why I give the volunteers the edge).

The pressure to interoperate in the healthcare space is growing constantly. Billing was the driving force for the past two decades. Exchanging data in a timely fashion between hospitals/clinics and payors drove innovation, and most big problems have been solved. However, on the clinical side, there is a lot left to do. Medical records need portability to support increasing geographic mobility, medical advancements, emergency services, and disaster management.

All these requirements at massive scale (national and international) necessitate either a common foundation or multiple foundations, tested and guaranteed to interoperate well (which is quite an expensive proposition). For that reason, using an open source, collaborative model for developing the foundation of healthcare technologies to be used at large scale by commercial (and non-commercial) entities sounds like a no-brainer.

There is a lot to learn from established open source organizations, the Apache Software Foundation in particular, whose projects are at the core of most successful commercial endeavours in their respective spaces. For example, in the integration space most current Enterprise Service Bus offerings (ESBs) are based on Apache Foundation projects like Apache ActiveMQ and CXF; the same is true for projects involving big data technologies like Apache Hadoop or Spark, and there are many, many other examples.

It's hard to tell if (or when) new open source foundations will appear and claim a leading role in healthcare. It would be interesting to see one created to scale an existing viable model, such as the one from Oroville Hospital using VistA. Or we could see OSEHRA shifting its focus and expanding its charter beyond just the US government space. Nevertheless, the successful foundation would keep a low barrier to entry for innovators, allowing them to incorporate and scale open source healthcare technologies into commercial products. Time will tell, but what's for certain is that we live in interesting times, and I am looking forward to massive innovation in healthcare in the near future. The time is ripe.