Open source is the current norm for developer collaboration and customer adoption in software. It is the foundation that enabled unicorns and cloud providers to build their services from the ground up. But that wasn’t always the case with open source, and it is changing and evolving again.
Open Source Eras and relative adoption trend lines
In this post, I will look at open source evolution broadly, try to analyze what are some of the triggers and enablers for the change, and where it might be heading next. Let’s start with the main open software development eras by summarizing the main trends and then focus on the big picture with an attempt to predict the future.
Free Software (1980)
The term “free software” is attributed to Richard Stallman around the 1980s for using it for the free-software movement. During these early days of computing, Richard started the GNU project in an effort to cultivate collaboration among the early hacker community and create a freedom-respecting operating system. He campaigns for software to be distributed in a manner such that its users receive the freedoms to use, study, distribute, and modify that software. This era set the origins of open source and more importantly the free software licenses (such as GPL) that flourish later.
At the time, the main software creators in the open were the individual hackers and in their view of the world, the software had to be free as speech and remain so. Free software grew because personal computers became more widely available to these hackers and they used CDs, floppy disks, and the early internet to distribute software and spread their ideology.
In this pre-internet era, manual distributions of software, supporting documentation, consulting services (installation, development), selling-exceptions were some of the popular monetization methods.
Open Source Software (2000)
The term “open source” was used by a group of people from the free-software movement around 2000. The motivation for this new term was to free itself from the ideological and confrontational connotations of the term “free software” and make it more appealing for the business world. The supporters of the open source movement stress the subtle difference from free software where free software requires any changes to be submitted to the original maker for redistribution, and any derivative software must also be distributed as free software. This new term set the beginning of a new movement and the forming of Open Source Initiative to educate and advance open source culture. The open source movement allowed smaller companies to play a more significant role in the software economy by giving them access to the software needed to compete in the global market. Before that, it was the larger corporations, the producers of the networks and hardware who had the power.
Open source sparked from the early hackers community but grew rapidly into open source businesses, enabled by software foundations, the internet, and the wider adoption of open source by companies of all sizes. The primary monetization mechanism for the open source software is through support and the open core models where additional accompanying value is created around the core open source project. While this open core (enabled by permissive licenses such as MIT, Apache) allows everybody to benefit from it, it is also its Achilles’ heel as we will see next.
Shared Source Software (2020)
Open source licenses give more freedom to the users, but they don’t give many advantages to the producers of the software. Many small projects with a handful of maintainers create huge economic value which ends up captured by other companies with better operational capabilities to monetize. This leads the maintainers of these projects to remain below the poverty line. Other companies hire open source maintainers as full-time employers and bet their company existence and brand into the success of their open source project. Yet they got disrupted and threatened by even larger hyperscale SaaS providers who have the scale to capture the economical value more efficiently and faster from the same projects.
This new economic reality started forcing individual maintainers and small companies to move their software away from business-friendly open source to other free software inspired derivative licenses and pursue dual-licensing models. This new family of licenses is not proprietary, but they don’t fit the open source definition either as they protect the trademark owner from the competition by discriminating against certain ways of software distributions such as SaaS. This transition of new and existing open source projects to non-open source licenses indicates the start of a new era. Keeping the source partially open is primarily for marketing and user adoption purposes rather than collaborative development and keeping software useful for everybody. This shared source software era is triggered by the existential threat of not being able to offer the software in a way demanded by consumers (as a SaaS) and efficiently capture economical value by the creators whether they are individual contributors or large companies with an open source business model.
Open source software eras and main characteristics
Protected by these new licenses, the enablers for the modern-day independent hackers are the powerful online services that allow them to offer good quality software through globally available automation tools based on git, build tools, software scanning, and distribution services, etc. These hackers can build enough critical mass of supporters through social media and are able to capture economical value through services such as Github sponsors, Patreon, Tidelift, and many others. The other group, the disrupted open source companies are transitioning to the SaaS based distribution of software as vertical cloud services on top of the hybrid cloud infrastructure to compete with cloud providers. This allows the creators of the software to offer their service on multiple clouds and at the same time align with the way users prefer to consume software, which is as a service.
What Will Software After Open Source Look Like?
The start of a new trend doesn’t indicate the end of the existing eras, but a new addition to the mix. Free and open source software will continue growing at a huge pace. At the same time, I believe we will see an acceleration of the trend towards the so-called shared source and source available licenses too. This will double down on the dual-licensing of smaller library projects by individual developers and the SaaS-based distribution of bigger projects. The open core and open source models will remain here, but the open core of the projects will get smaller and smaller, practically useless for the competitors. We will see projects starting as open source during bootstrapping and initial adoption phases, and then transition to source available licenses when threatened by more operationally mature competitors. Unfortunately, this initial phase of uncertainty and adaptation in the shared source era will limit collaboration among competitors and demonstrate the importance of open governance and open funding through neutral software foundations or decentralized technologies.
Then we will see cycles repeating and independent hackers flourish again, innovating as in the free software era. But this time they will be better equipped with better infrastructure to support their livelihood as independent small businesses of one. They will start projects in the open to scratch their itch, but quickly turn them into businesses or let them die. They will be less ideological, and more practical. These independent hackers will not need to be part of the traditional horizontal software companies that bundle engineering, marketing, sales, support, education, etc to be successful in the software business. Instead, they will be able to consume unbundled vertical online services and deliver enterprise-grade software. We will see a rise in the tools and platforms that offer reliable project governance without joining a foundation or consortium. Independent software builders can use decentralized infrastructure, tokenize their projects, and customize the governance through on-chain community voting. The economical and governance aspects of the projects will be merged with the source code and licenses into a holistic entity enabled by blockchain technology and create opportunities for individual hackers to create million-dollar companies.
The infrastructure for independent techies will not be only for the software builders but for the whole ecosystem. Creating software is not enough, it has to go through the full pipeline of budgeting, building, marketing, hosting, sales, support in order to grow and remain sustainable. Speculators will put money into project tokens to help bootstrap projects and gain returns. Developers will build. Indies will create niche services complementing larger projects. Subject matter experts will provide consulting services and online training, and bounty hackers will hunt for ad-hoc work. Sometimes all of it will be driven by a single person, and sometimes a whole decentralized ecosystem forming around a project without the dominance of a central business entity. This will take a generation of software builders…
At the beginning of the open source era, Eric S. Raymond described a decentralized software development process called The Cathedral and The Bazaar. This era proved that the bazaar is the superior software development model. But at the same time, this era also showed us the limitations and the narrow mindedness of this model when it is not accompanied by a holistic governance and monetization view. The next era will improve on the same decentralized development principles by incorporating decentralized monetization and governance too. This will take us the full cycle of Decentralized and Sustainable Open Software nirvana.