The UNICEF Innovation Fund invests exclusively in Open Source technology – with today’s rapidly evolving innovation landscape, Open Source software, hardware, data, and content not only create value and generate revenue, but also ensure greater collaboration and impact. This reflection is a look back at Open Source activity and participation stemming from the UNICEF Ventures Team from June 2020 to date (July 2021).

By the end of this article, you will have a better understanding of the evolving and forward-thinking approach to Open Source taken by the UNICEF Office of Innovation.

This article looks at a few aspects of Open Source engagement at the Innovation Fund:

1. Support models
   1. Legal & policy
   2. Building and leveraging from the community
2. Case study: Cloudline and upstream engagement

New to Open Source at UNICEF? 🔗

If you are hearing about the UNICEF Open Source Mentorship programme for the first time, check out this introduction post to get important context behind why UNICEF invests in Open Source mentorship and support for Innovation Fund portfolio teams:

https://jwfblog.wpenginepowered.com/2021/07/unicef-open-source-mentorship/

Innovation Fund support models 🔗

Start-ups receive both funding and customized mentorship during their 12-month investment period with the UNICEF Innovation Fund. Since the Innovation Fund launched in 2016, the Innovation Fund team noted the impact of hands-on guidance to help start-ups understand how to work open and lead open. Start-ups receiving Open Source support were better equipped to develop sustainable business models that made Open Source intellectual property work with them instead of against them.

By the end of the 12-month investment cycle, graduating companies achieve the following milestones in place with their projects:

• Solid understanding of Open Source licenses and different business models depending on a permissive or copyleft strategy.
• Laying foundations for growing or participating in friendly, inclusive communities.
• Documentation site to showcase their Open Source work, and how to get involved.
• Continuous Integration pipeline to test new changes in the code-base before they are added, and avoid common human errors.

Across these milestones, three themes of support have emerged as as most essential in the past year: legal & policy, building and leveraging from the community, and building in the open.

Legal & policy 🔗

The UNICEF Innovation Fund is unique in Venture Capital on its open-first investment strategy. Start-ups receiving UNICEF funding are obligated to either create their own Open Source works or contribute to existing Open Source works. This leads start-up teams to ask more questions: how does this work in a legal sense? How do you establish a sustainable business model with Open Source dependencies and the different types of license models that exist (e.g. permissive and copyleft)? These are the questions that start-ups receive tailored guidance on from the Open Source Mentorship programme.

How did the Open Source Mentorship programme support both start-ups and UNICEF Country Offices receiving funding from the UNICEF Innovation Fund? The programme improved business and legal resources and referrals for managing Open Source intellectual property. This was primarily done by creating a self-serve knowledgebase of legal, governance, and tech policy: the UNICEF Open Source Inventory. The Inventory was created in close consultation with start-ups receiving UNICEF funding, leading industry experts in areas such as Open Hardware and open design, and other existing Open Source communities of practice.

Building and leveraging from the community 🔗

Over the past year, we also saw more upstream collaboration. UNICEF Innovation Fund start-ups collaborated on existing projects like the PX4 Drone auto-pilot software instead of reinventing the wheel themselves. This enabled the work done with UNICEF funding to go further and impact the wider ecosystem, instead of an individual project with highly-specific use cases.

Furthermore, as the Innovation Fund portfolio continues to grow, new inductees in the Open Source Mentorship programme are able to build on top of work done by previous portfolio companies. This accelerates the rate of development for the new inductees and gives them a model of success to look towards during their engagement with UNICEF.

For example, two companies from the 2019 Drones cohort created documentation websites that were used as models for the 2020 Generation Unlimited cohort. qAIRa from Perú and Rentadrone from Chile created documentation websites using the popular toolchain Docusaurus, created by Facebook’s Open Source Program Office. The work of qAIRa and Rentadrone was leveraged as models by VRapeutic and I-STEM in the 2020 Generation Unlimited cohort. Using graduated companies as models accelerated both technical and content development for the new teams when they had relatable models to use in building their own Open Source documentation websites.

Innovation Fund case study: Cloudline and upstream engagement 🔗

Cloudline Africa (South Africa) operates small-scale autonomous airships that have longer endurance and range than current commercial drones; their solution will help deliver medical supplies to hard-to-reach communities and reduce operational costs in the last-mile.

Four highlights from Cloudline’s period with the Innovation Fund are below:

• The team launched the airship with a payload capacity of 10 kgs with a 50 km range.  
• Full endurance capability (40km/h) along with automated waypoint flying. 
• Contributed flight control software to the PX4 upstream community by introducing a new mode of aircraft to the community in the form of airships.  
• Cloudline received the Fast Company South Africa Most Innovative Company Award 

“Our greatest triumph of the year was the flight of our pre-production airship. Capable of a 10kg useful payload and a 50km range, this aircraft is ideally suited to medical logistics and other high-value scheduled deliveries.” — Spencer Horne, Co-Founder

Going upstream 🔗

Contributions made to another popular upstream community was another major highlight of the Open Source Mentorship programme. Furthermore, the contributions were not casual “drive-through” contributions; they were consistent and focused broadly around the ecosystem instead of a single repository of source code.

During the mentorship programme, Cloudline added 61 commits across four repositories. The summary of their contributions were adding a new vehicle type to the PX4 auto-pilot software: airships. The team first opened a discussion on the PX4 community forum. After getting feedback from a developer, they proceeded to add the Cloudline airship into the upstream software. This was followed by documentation additions with controller diagrams and simulation instructions. Through their contributions, a wider network of developers can simulate and interact with airships similar to the one created by Cloudline.

But the contributions did not stop at code. Cloudline developer Anton Erasmus delivered a deep dive presentation at the PX4 2020 virtual contributor summit. He explained to other developers about how to leverage the multicopter control system, as used in the Cloudline airships and other popular drone devices. This outreach step is important in growing awareness in the existing community for new introductions to the common software, and how both Cloudline and the PX4 community could help each other in achieving their similar yet different goals.

What is next for the Innovation Fund? 🔗

What does the next year ahead look like for UNICEF’s Open Source Mentorship programme? Stay tuned for the final post in this series, which will explore on-going work to standardize the mentorship curriculum, detail how Innovation Fund companies are being coached in becoming Digital Public Goods, and the research and development for an Open Source Program Office at the UNICEF Office of Innovation in 2022.

Special thanks for content and data collected in this post: Sanna Bedi, Zenani Orengo

Many have started on this same path before. Several communities form a committee as a governance model for important decision-making. Usually committee membership is chosen through an election process. Paid employees, or sometimes, members of the community comprise the elected committee membership. This meritocratic approach is believed to bring in diverse representation and participation of highly-engaged people. After all, who better to represent contributors of a project than a committee of folks elected by their own peers?

Sometimes, committees do accomplish this lofty goal. My argument is that sometimes they don’t – especially if your committees are designed in a way to disable participation.

Context brief: what is a committee? 🔗

Frequently in this post, I refer to committees. But what are committees? I see a committee as a I see a committee as holding the following characteristics:

• Fixed group of individuals charged with important decision-making privileges
• Appointed or elected members with fixed term periods (i.e. an end date)
• May perform their work in a public and transparent way

Challenges of a committee 🔗

If designing a community for participation and engagement, a committee can do the opposite by pushing people away. It can be difficult for non-members to participate in important decisions. When building the foundation of a community on volunteerism, expecting others to give time in huge quantities is a false expectation. An active, long-term commitment as a committee member may be a big ask. Yet even if an individual wants to contribute, their company may not support such policy. So, this person is unable to contribute fully in the committee. Therefore, the opportunity is lost to include their voice as a representative of a larger community.

Furthermore, a committee depends on the engagement of its members to be effective. Committees are limited by the amount of time individual members actively contribute. Committees lose their effectiveness when:

1. Individual committee members practice poor time management, or are simply overloaded with too many responsibilities
2. Inclusion of others with valuable perspectives have no pathway to being heard or represented, unless they are on the committee

Committee members participate for a fixed amount of time as regular participants. This can be good for a healthy turnover rate, but it becomes bad when the same people are running over and over again. Often described as burnout!

What is a better design for community engagement? 🔗

A fatal flaw in community management is being too hands-off or too hands-on from a corporate context. I look back at 2018 in the difference of roles in Fedora Mindshare vs. Fedora CommOps. Red Hat strives for participation beyond paid Red Hat employees, yet the volunteer-driven community struggles at times for participation of any Red Hat employee.

The Mindshare Committee is the community body that leverages power in the community. These are tasks that could have been designed by CommOps too. I think the format and spirit of CommOps encourages collaboration and invitation to contribute. On the other hand, if you are not an elected or appointed member of the Mindshare Committee, there is not much in the ways of contributing. Even if that is more a belief than a fixed rule.

2020 saw the launch of a formalized Open Source Mentorship programme for the UNICEF Innovation Fund, built up on two years of work from RIT LibreCorps expertise and consulting.

The Open Source Mentorship programme includes five modules about Open Source intellectual property and communities delivered across twelve months. UNICEF grantees are matched with an experienced Open Source Mentor to guide them through the modules. The mentorship takes an interactive, guided approach to understanding the unique context that each team and product exist within. The assigned Open Source Mentor provides specialized advice and training:

• Tailored feedback based on business models
• Existing local user communities
• Best practices for collaborating together with others on similar challenges.

The geographic diversity in the UNICEF Open Source Mentorship programme is unusual for technology incubators or accelerator programs. All funded projects come from UNICEF programme countries. The UNICEF Innovation Fund provides equity-free funding for Open Source solutions from local innovators and entrepreneurs solving local problems. To date, the Innovation Fund has invested in teams from over 57 countries. Argentina, India, Iran, Kenya, Mexico, Nepal, and Rwanda represent the most recent incoming cohort in July 2021.

But why does this kind of work matter to UNICEF or the United Nations?

UN Roadmap for Digital Cooperation 🔗

In June 2020, the United Nations Secretary-General released the UN Roadmap for Digital Cooperation, a call for action and a vision with several key considerations:

• An Inclusive Digital Economy and Society
• Human and Institutional Capacity
• Human Rights and Human Agency
• Trust, Security and Stability
• Global Digital Cooperation

The report explores the impact technology has in each key area:

“Digital technology does not exist in a vacuum – it has enormous potential for positive change, but can also reinforce and magnify existing fault lines and worsen economic and other inequalities. In 2019, close to 87 per cent of individuals in developed countries used the Internet, compared with only 19 per cent in the least developed countries.”

UN Roadmap for Digital Cooperation, June 2020

Open Source Mentorship enables digital cooperation. 🔗

Therefore, the conception and development of Open Source Mentorship is inspired by the Roadmap, in four key aspects:

1. Digital Public Goods: To encourage knowledge transfer and collaboration on practical, everyday advice on building technology projects in line with the Digital Public Good Standard.
2. Digital Capacity-Building: Empowering others with the skills and information they need to be bringers of Open Source change and innovation in a disruptive and competitive Venture Capital ecosystem.
3. Digital Trust and Security: Providing a holistic view of Open Source intellectual property that is influenced by, but not bound by, quarterly earnings reports and share prices. Understanding the commitments of Open Source licenses and technology, and how trust is earned and lost in Open Source products and communities.
4. Global Digital Cooperation: Designing and structuring communities to be healthy and collaborative in nature, united in resolving common challenges and problems.

Sustain Summit events are led by a facilitator. There are no keynotes, talks, or sponsor demos. Your undivided attention is required. Phones and laptops should not be used throughout the day and you will be asked to put devices away if they are a distraction to you or anyone else.

When we talk about sustainability, we are talking both and equally about the sustainability of resources and the sustainability of its people. We hope you can join us for the conversation.

sustainoss.org/summit-2020/

This is my second time attending Sustain OSS (see my 2018 event report). I attended on behalf of RIT LibreCorps to represent the sustainability efforts at the RIT FOSS@MAGIC initiative, but also to represent myself as an individual and sustainer in the open source movement. For Sustain OSS 2020, I arrived hoping to learn more about community-first governance models. I left with a lot of notes and the first blueprints for Principles of Authentic Participation.

Event reports take many forms. Since Sustain is structured in a unique format, my event report is structured as follows:

• At a glance: structure and key takeaways: High-level overview of what the conference is like and the biggest ideas on my mind at the end of the day
• Community-first governance: I came ready to explore this idea, and had a unique conversation about citizen assemblies
• Principles of Authentic Participation: Impromptu session I facilitated, and thus spent most of my time focused on

At a glance: structure and key takeaways 🔗

If you’re here for the quick overview, this is it.

I loved Sustain OSS 2020 because it is a unique collection of people from various backgrounds in the Free/Open Source movement. Both old and new folks, software engineers and designers, open source program office folks and the FOSS lawyers, all together in one room. Perhaps the best part for me is leaving with a sense of empowerment and connection to a bigger movement of people.

Speed breakout sessions 🔗

The first half of Sustain 2020 started with speed breakout sessions led by many facilitators. There were around twelve small discussion groups focused on specific topics. You could spend 4-5 minutes at six topic groups. Choosing was hard! These speed sessions are primers on what to spend the second half of the day focused on, in a more detailed discussion.

I attended these sessions:

1. Diversity and inclusion
2. Open source movement in Africa
3. Minimum Viable Product: Good governance models
4. Documentation
5. 20 years of sustainability
6. Models for corporate accountability as open source community members / what does it mean to be a corporate member of open source communities?

Key takeaways 🔗

I came up with four key takeaways from Sustain OSS 2020 as a whole (not including the detailed sections further below):

Inclusion is local 🔗

When reaching out to new areas and demographics, include local community leaders. This is to say, if you are organizing communities in Asia or Africa, the success of your outreach campaign depends on your ability to enable and include existing community leaders in these regions. Local perspective is required for authentic grassroots success.

Rules for revising rules 🔗

When defining community governance or policy, expect change. So, include ways to change the rules later when the world changes around your governance or policy.

Generalists transition to specialists 🔗

In the early phases of an organization or project, community members are often generalists. Fewer people wear many “hats.” But context-switching has a cost. As the organization/project grows, defined roles become more important. Defined roles avoid everyone doing everything.

Designers focus first on design. Developers focus first on code. By specializing, you maximize the potentialities of what your team brings to the table.

We got the power 🔗

There was a breakout group about ethics in Free Software. Two emerging themes were creating ethics review boards at organizations and the power of labor organizing. Some suggested normalizing ethics training in employee on-boarding.

Most notably, there was a highlighted need for safer spaces for labor organizing discussions. Labor organizing comes at a high personal cost for many.

Community-first governance models 🔗

I came to Sustain OSS 2020 ready to explore community-first governance models. In December, I published a blog post on why Free Software is still not on activists’ agendas. Free Software outreach often emphasizes technology, not people or ethics. We focus on technology so much, we forget why this movement began in the first place. So, in a world where corporate interests in a project often conflict with interests of grassroots communities, I wanted to know what “community-first governance” really means.

For context, I consult with humanitarian organizations and non-profits that want to build community around their open source projects. But from personal experience, I realize community stakeholders need input to decision-making if the community is going to stick around.

Citizen assemblies in open source? 🔗

While I didn’t explore this in a session, I did have an interesting conversation with Xavier about citizen assemblies and how open source communities might be modeled after them. Xavier explained citizen assemblies to me as a form of hyper-local representation in policy-making in regional governments. For example, U.S. citizens are obligated to serve on jury duty, or a jury with your peers on a court hearing. Similarly for citizen assemblies, individuals are selected at random based on different demographic factors. Those selected serve a period of time on a local legislative body.

In open source projects divided across different contexts, think about the demographics represented in our communities. What might the demographic factors look like?

• In internationally-dispersed communities, it might be approximate region or nationality.
• In corporate-focused projects, it might be diversity of employers.

Citizen assemblies in open source could mean a group of contributors are selected at random to serve on the executive decision-making body of a community. The decision to serve is always optional. If a selected person declines, another person is selected.

I think this is a crafty way to address a cultural divide often present in Global South communities. Many contributors subconsciously look for an invitation to contribute. Being explicit by selecting an eligible, qualified candidate actively includes perspectives not commonly represented. It could also be a way to get around imposter syndrome and self-(non)selection in communities that use committee governance models.

I am still sitting with this idea. But I think Xavier’s idea of applying citizen assemblies to open source communities is valid. I’m interested in exploring what this model could look like in practice when the opportunity for experimentation presents itself.

Principles of Authentic Participation 🔗

I attended a session about corporate accountability in open source contributions. It evolved into a longer discussion about Principles of Authentic Participation. This week, I launched a wider discussion about this on the Sustain OSS Discourse forums, so instead of repeating myself in this post, read the Discourse thread and leave your thoughts there.

However, for the reader’s convenience, the six principles we drafted are below:

Authentic Participation Starts Early. 🔗

This came out of discussions about organizations showing up with mature, fully-baked contributions over which the community had no input.

Authentic Participation Puts The Community First. 🔗

This reflected the general consensus that when an organization and the community want different things, the community needs to come first.

Authentic Participation Starts With Listening. 🔗

This was Duane’s reflection of some comments about folks showing up to projects with no historical context and telling them everything they were doing wrong.

Authentic Participation Has Transparent Motivations. 🔗

Without a shared understanding of the motivations, it’s impossible to resolve differences of opinion effectively. No hidden motives.

Authentic Participation Enforces Respectful Behavior. 🔗

This is about organizations taking responsibility for the behavior of their representatives.

Authentic Participation Ends Gracefully. 🔗

No sudden withdrawal of resources without notification and an exit plan. Clear documentation that would allow the community to pick up projects when a company decides to withdraw support.

Thanks folx! 🔗

To wrap up this Sustain OSS report, a few thank-yous are in order:

• Stephen Jacobs: For always being supportive for yet another trip abroad and helping me push my career forward in a number of ways (and footing the bill!)
• Mike Nolan: My co-conspirator, partner in FOSS, and comrade in arms
• The Principles of Authentic Participation group: I didn’t plan to facilitate at Sustain and I was nervous about it, but you all were wonderful. We had a fruitful discussion and I’m looking forward to the follow-up.

Sustain OSS 2020 continues to give me a lot to think about and consider. I’m fortunate to have attended. I hope this event report gives additional visibility to some of the conversations held in Brussels this year.

Where Byfield was accurate 🔗

While I don’t agree with all of Byfield’s sentiments, he identified some key challenges that still hold truth today: a predisposition to focus on differences and not similarities, an outreach approach centered on ethics and not software, and the importance of opportunities for intersectional interaction.

Predisposition towards difference 🔗

First, Byfield notes the differing age groups of the activist communities and the tendency for viewing others by their differences first, not their similarities. He seems to attribute the tendency to view others by differences first as a characteristic of older generations; however, this is not necessarily the full truth.

As a member of the young activist community, this explanation is too simplistic of the underlying cause. There is also a political motivation by established power to sow division among the population of a nation-state. It makes community organizing more difficult and presents diversity as an issue to “solve” instead of a source of greater unity and common strength.

This is exemplified by the social media algorithms of today that reward sensational content (judged on likes, views, clicks, or other user feedback) and share it widely across a huge platform. In 2006, it was difficult to imagine the relationship social media would have in the lives of an everyday person; today, a great deal of social power is granted to those who understand how to leverage social media, either for good (e.g. social activism) or harm (e.g. deceptively persuading large parts of a nation-state’s population leading up to a national election).

The politics of division are within the fabric of our political systems; this is a challenge for modern-day activism and community organizing to overcome. In identifying this as a challenge, Byfield is correct that a differences-first approach makes it harder to share and spread the importance of FOSS in other activism circles, especially as technology becomes an increasingly relevant way of how we experience our lives and how our systems of law and justice are enforced.

Outreach on ethics, not software 🔗

Second, Byfield suggests an ethics-based approach to outreach is more effective than a software-based approach. This is also correctly noted, even if perhaps overemphasized. The jargon and language of the technology world is not accessible to the large majority of the global population. While some degree of technology literacy might be expected in some populations, much of the existing FOSS community is deeply rooted in technology. Sometimes this limited perspective is counterproductive.

This revisits the rebranding of “Free Software” as “open source” in 1997. For many subsets of the wider open source community in 2019, the default approach to open source software is merely a secondary thought for how to collaboratively work on technology. This is part of the outcome of the Open Source Initiative’s gamble in 1997 by beginning to emphasize the business sensibility and practicality of open source, and de-emphasize the social roots of Free Software (or rather, try and position itself as some sort of translator between these two “worlds”, as if they cannot be spoken of together in the same room).

As such, those who work on open source software projects are not necessarily predisposed to assume the role of an activist. Truly if Free Software is to take root outside of technology, then those who see the ethical values of Free Software need to better organize and promote the values of FOSS externally. This will contribute to the diversity of Free Software activism by helping non-technology activists add FOSS as a tool to their existing work.

Intersectional movement building is the future 🔗

Thirdly and finally, and perhaps most importantly, Byfield suggests the importance of intersectional interactions between Free Software communities and other activist communities. This is a fundamental requirement for the growth of Free Software as a social movement. Those of us in Free Software see the world around us informed by a background informed by technology; this background is emphasized in a world that is generating new, advanced technology at an unprecedented rate. However, while software and technology are important parts of the world around us, they are not the world around us. They are one part of a greater picture of fighting for a common good and welfare for all people. There are others in similar niches who have a deep understanding of their problem space and how they want to approach a challenge.

“One group may be working against child poverty, another for recycling, but the people in these organizations can almost be transferred from one to the next.”

Peter T. Brown, Free Software Foundation Executive Director (2006)

Just like a healthy garden, cross-pollination of these niches is vital to help others understand how we can help each other in accomplishing our mutual goals (this also feeds into why the politics of division explained above is so pervasive and difficult). Bringing Free Software technologists to activist communities where there is not an overwhelming Free Software background (and vice versa) is vital to building an intersectional social movement that strengthens the social impact of Free Software, not just open source.

Where Byfield didn’t go far enough 🔗

Byfield made one assumption on how activists have “their own share of insularity” and that the presence of connections between two movements does not mean they could immediately connect their existing beliefs with new ones. Fioretti’s challenge was in others understanding why they should listen to him; there was a lack of foundational knowledge of open source and technology that is normally assumed of someone who works as a software engineer.

Indeed, attention is a currency in the world of an activist. It is not enough for a FOSS advocate to expect others to listen to you on an appeal of technology. Part of the work in sharing is understanding who you are sharing with; if FOSS wants to take deeper roots in the activist community, it needs to understand the backgrounds of activist communities and be creative in how to appeal the mission of FOSS to the mission of their work. Where you can build in-roads together with others through common initiatives is the beginning of grassroots community organizing. So, while Byfield is right that there is an almost competitive nature of ideas in activism, it is not enough to write insularity off as a fixed aspect of nature. To not acknowledge this is to deny the influence of capitalist power structures in the humanitarian sector as they pertain to sustainable funding.

What are today’s challenges? 🔗

Some of today’s challenges are about inclusion and power.

Inclusion builds power 🔗

Diversity and inclusion (D&I) are important but poorly understood; not only are D&I about including people of different identities in technology, but also people with backgrounds outside of technology. FOSS stands to benefit by including more people who do not necessarily have a strong technology or engineering background. The goal is to inspire different perspectives to contribute in meaningful ways to build sustainable technology.

Instead of seeing diversity and inclusion initiatives as problematic or unneeded, D&I groups in FOSS communities stand to be the most effective people at building community and influence.

Power and governance 🔗

In the activist / humanitarian / non-profit world, there is a power struggle for sustainability as it pertains to funding. Funding models in non-profit work (usually sustained by grants, sponsors, and donors) encourage solutions that get funded, not necessarily solve problems the most effective way. Many organizations struggle with how to achieve sustainable funding without being so dependent on the expiration date of a grant’s funding.

We need more representative governance models in open source communities that reflect the interests of the communities around them, not necessarily an individual, a company, or group of companies. Building governance models that empower people within a community to make decisions and reduce the corrosive influence of money from humanitarian work.

Where do we go from here? 🔗

This blog post is an active reflection of my own thoughts and perspectives of Free Software, activism, and humanitarian work. If you are interested in pushing this conversation further, find me in Brussels, Belgium for any of the following three conferences and let’s chat further:

• 30 January 2020: Sustain Summit
• 31 January 2020: CHAOSScon
• 1-2 February 2020: FOSDEM

If you want to discuss this further, you can also drop a line in our online discussion community, fossrit.community.

Photo by Walid Berrazeg on Unsplash