Foundation/AGM20/Election to Board/Answers and manifestos/Tobias Knerr
Your OSM activities
I signed up to OSM in 2008 as Tordanik. I no longer remember where I first discovered the project – it may have been an article in a prominent German computer magazine. At the time, I already was a strong supporter of free software and open content, though, having previously contributed to Wikipedia and Commons, so OSM was right up my alley.
My first edits to OSM involved mapping my home town and university town. For big chunks of them, I did that pretty much from scratch, using local knowledge plus a GPS logger which I had purchased for this purpose. Today, my local area is still the focus of any mapping I do, although my mapping interests have become more specialized (such as 3d mapping, and micro-mapping in general). I've also spent quite a lot of time improving the OSM wiki, and helping out people on the forum, the mailing lists and on help.osm.org.
Throughout the years, I've had the opportunity to explore many facets of our amazing project. Together with the great folks from my local mapper meetup, I've organized mapping parties, taught newbies to use JOSM, presented OSM to students as well as the wider open source community and at industry events, and arranged a cooperation with the town's administration. We hosted the 2017 FOSSGIS conference (the de facto "SotM Germany", with a couple hundred visitors), and I've been a frequent speaker at several OSM events. Some of you may also be familiar with OSM2World, an open-source 3d rendering tool that I maintain as a volunteer developer.
As for foundation matters: I'm a member of the Communication Working Group, doing mostly English-German translation work for the OSMF blog. Having served as mentor and co-admin for OSM's Google Summer of Code participation since 2015, I additionally joined the Engineering Working Group when it was entrusted with managing GSoC. And of course, I was elected to the board of the OSM Foundation in 2018. It has been an intense and exciting time, and I believe we've made good progress as a board and a project – I'll go into more detail in other questions and my manifesto. With your support, I would be happy to serve another term!
Why you want to be a board member?
When I joined the board two years ago, I made it a priority to strengthen the influence of volunteers in the OSMF. In a project otherwise designed to be distributed and do-ocratic, the OSMF and its board are outliers. They are uniquely positioned to do a lot of good – I'm discussing my ideas for this in other answers – but they are also potentially vulnerable to takeover attempts, an outsized influence of organized groups, and other mechanisms that would upset the balance of power in the project.
I'm happy that we were able to implement many of the immediate measures I had outlined in my 2018 manifesto: Offering OSMF membership to active contributors. Introducing Conflict of Interest rules for the OSMF board. Linking OSMF membership to OSM accounts and paving the way for transparent membership lists based on OSM user names. And, at the upcoming AGM, asking for a mandate for additional protections against takeover attempts and companies' control over their employees' actions within the OSMF.
Still, this is very much an ongoing concern. Some steps are already in the board's pipeline, such as defining membership prerequisites and extending our conflict of interest rules to working groups. Other changes initiated by the board during my current term, especially the creation of additional paid positions within the foundation, will also pose new challenges in this area: We must avoid becoming dependent on any particular source of funding, and ensure that volunteers continue to be first-class citizens in an organization that has employees and contractors working for it. Making progress and adapting to changing technological and social circumstances without abandoning our core values will always be a balance act, and there's a lot of work to be done!
There's no doubt that board work takes a considerable amount of time and energy. Serving on the board for these past two years has occasionally been challenging, and it has definitely put a limit on the time I was able to dedicate to my other OSM activities, such as working group participation, mapping, and software development. However, I believe it's important for me to continue spending at least some time in those areas – not the least to stay grounded in the practical realities of the project. If I'm re-elected, I plan to continue this overall approach without major changes.
Do you have any previous relevant experience?
Having been an OSMF board member for two years surely counts as relevant experience. :)
More generally, I believe that there are some traits all board members should share: Love for the OpenStreetMap project, commitment to its values, and the ability to constructively work together in the interest of the OSM community as a whole. Because I've contributed to OSM for over a decade in various roles, I'm confident that I have a decent understanding of how OSM works.
My professional experience is as a researcher and software developer. The mode of collaboration I enjoy most is teamwork among equals, built on trust, competence and mutual respect more so than formal authority. I'm not a manager, nor do I want to be. I highly value what my board colleagues with leadership experience in traditional organizations bring to the table, but at the same time, I don't believe that this is the only perspective which should be represented on the foundation board. OSM's community and core values set it apart, and we should not unthinkingly copy patterns found in hierarchical organizations.
Transparency: Conflicts of interest
I'm participating in OSM as a volunteer in my spare time. For my main source of income, I work as a researcher at the University of Passau, mostly on projects related to the automotive industry and self-driving cars. Map data does play a role here, so although we aren't involved with OSM, some conflicts of interest are possible.
I'm also a member of the OSMF local chapter in Germany. Otherwise, I'm not affiliated with any of the organizations which do work in the OSM ecosystem. My communications and actions as a board member are not limited by any relevant contracts.
What to do with the face to face meeting in Corona times?
Due to the pandemic, we decided to replace the board's face to face meeting in spring 2020 with a video conference. I'm convinced this was the right call, and that the first such meeting after the election should be virtual as well.
My first-hand impression as a board member is that this year's virtual events have been, for the most part, just as productive as the previous face to face meetings. And the lower barrier for organizing a virtual event that has made it easy for us to hold a second such meeting in 2020.
So although I would greatly enjoy meeting my fellow board members in person (again), I do think we can learn from this year's experiences: Given the greater cost and environmental impact, as well as the logistics that will only grow more challenging as the board becomes more geographically diverse, not every such meeting needs to be in person - even once the threat of COVID19 is behind us. (Fingers crossed!)
What's the use of the OSMF ?
The most fundamental purpose of the OSMF is to keep the lights on: To make sure the technical and legal infrastructure OSM depends on is running smoothly. There needs to be computer hardware somewhere for our map data to live on, for example, and the OSMF was set up to be the entity dealing with those necessities.
But there are ways in which the OSMF can and does serve the OSM community beyond that baseline:
- Stand up for the community's interests in dealings with other organizations. At times, a loosely organized collection of individuals can have a hard time making its legitimate concerns heard when facing more coordinated groups such as large corporations. Whether it's, say, an organized editing campaign gone wrong or missing attribution, the OSMF is often in a stronger position to communicate and enforce the community's expectations.
- Fund software infrastructure. Most software development in the OSM ecosystem is handled by separate groups, including a considerable number of companies building products on top of OSM data. This vibrant software ecosystem is one of our greatest assets! However, not every software project enjoys a viable business model. There are tools such as editors, open source libraries, and our API and website, which make OSM better for everyone but cannot easily capture that value. The OSMF can help make these projects sustainable. For you, this could mean that the OSMF funds the development of new features e.g. in your favorite OSM editor.
- Help contributors connect with each other. It's important to have shared spaces where OSM's global community can interact, and the OSMF is in a good position to provide these. Most of you are already using some of our channels to talk to other members of the community and stay informed about what's going on. Going forward, I hope the OSMF will be able to offer even more attractive platforms, and better tools for local chapters and communities to organize themselves.
Of course, the above is not an exhaustive list. There are many other examples, such as our recently launched Microgrants program. Throughout all of this, the guiding principle is that the OSMF supports, but does not control, the OSM project. The foundation wants to help the project succeed, but what success means is defined by the community.
What will you do to build a worldwide community of mappers?
The foundation has seen substantial progress in recent years: In 2020, we were able to welcome the first OSMF local chapters from North and South America, Oceania and Africa. Zero-cost membership for active contributors has been successfully introduced and has lead to a signficant influx of members in some parts of the world that were previously under-represented – and a community of mappers around the globe is a direct prerequisite for creating a map of the world from local knowledge. Most importantly, this development isn't limited to just the foundation: Map contributions, which are ultimately a far more important metric than foundation membership, show an encouraging upward trend around the world.
So even though there is still a long way ahead, I believe things are going pretty well, and I would not advocate for drastic course corrections. Instead, there are many smaller improvements that the foundation can make for the benefit of communities around the world. These include:
- Supporting and giving visibility to local groups and events, both for outreach and to foster closer ties between contributors who are looking for this kind of experience. After the pandemic, this would also include real-world meetups, mapping parties and conferences.
- Attractive and open online communication tools, such as modern forum and chat platforms, offered by the OSMF for all language communities, building on the work of OWG and the FOSS Committee.
- Through the Communications Working Group, putting a spotlight on OSM's success stories around the world, and improving the discoverability of community content on the OSM website.
- Assisting our local chapters, who are best positioned to meet the specific needs of their national and regional communities, actively involving them tasks such as moderation, and welcoming additional chapters into the fold.
What will you do to encourage more women leaders in OSM working groups and governance?
I feel that any significant and sustainable change in our community's demographics has to start with the volunteer community – with more women donating their spare time to build an open map of the world. It certainly would be a huge benefit for OSM to unlock this largely untapped potential of contributors. But honestly, I don't have any revolutionary ideas on how to achieve this, especially as a look at similar projects suggests that the causes may not be unique to OSM. Nevertheless, 2020 has seen some attempts by the board to improve the situation. The adoption of the Diversity Statement was an important symbolic step in this regard. And eventually, this year's creation of the Diversity and Inclusion Special Committee, which we plan to reinvigorate after a slow start, may result in more tangible improvements. At the very least, I hope it would lead to a better understanding of the situation, including some of the less frequently studied dimensions of diversity.
Should OSMF accept funding/donations by companies or organisations which do not want to be disclosed to the public?
No. Withholding the donor's identity from the membership would be incompatible with our commitment to transparency, and the foundation has rejected such offers in the past. We generally welcome corporate donations, but community oversight is an essential part of the protections against undue influence over the project.
Developing the OSM editor that will be many people's first mapping experience is not an easy position to be in. For the most part, I believe iD is doing an absolutely impressive job in offering a polished interface and a gentle learning curve, which are important virtues for the default editor on osm.org.
That being said, not all of the prominent controversies were an inevitable result of iD's role or the genuine ambiguities and unresolved problems OSM tagging is suffering from. Some could have been avoided by sticking to tagging backed by sufficient community consensus, or by a less divisive communication style. And not all controversies were tagging-related, either. For example, I find it unfortunate that the Community Index still recommends proprietary social media channels over OSM's own platforms unless that behavior is explicitly overriden by a local community, and that iD is relying on Facebook for brand images despite the privacy implications.
Now, I want to emphasize that controversies along those lines would not justify OSMF involvement for most software projects in the OSM ecosystem. You do not need to cooperate with the foundation to develop software which uses or creates OSM data. And even with financial support by the foundation, we're explicitly looking for people and projects capable of producing results without close supervision.
Instead, I'd say the key reasons for OSMF to pay particular attention to iD (as with the software dispute resolution panel) are its position as the default editor – a privilege ultimately granted by the OSMF, and a notable difference compared to tools that mappers must actively seek out – and its inclusion on the osm.org website which exposes the OSMF to legal concerns such as copyright or privacy. More generally, there's also a certain level of responsibility that comes with write access to the editing API, especially with editors increasingly abstracting tagging decisions away from the mapper.
How many paid staff should the OSMF have and why?
Over the past months, the OSMF board has substantially increased the amount we spend on paid work. While the absolute number of paid positions is still small compared to some other organizations, this nevertheless marks a significant change. The potential benefits for the continued stability of the OSM platform are evident, and experience shows that major projects from the open source and open data worlds tend to eventually come to rely on paid work to an extent (although professional contributors need not necessarily be employed by a single organization). Important work is sometimes neither fun nor inherently profitable, and payment can ensure that it still gets done reliably.
At the same time, there are serious risks involved with creating paid positions. Increased spending makes the OSMF dependent on the goodwill and financial support of wealthy backers. The group(s) within the OSMF who decide on fund allocation wield a substantial amount influence, and so do the people holding paid positions themselves, through access to core infrastructure and as a consequence of spending far more time on shaping the project than others possibly can. Compounding all this is the difficulty in going back to volunteer work once we've created a paid position for a particular task.
Therefore, I believe that the foundation should limit itself to a relatively small number of employees or long-term contractors, who would be working primarily on our core systems and critical administrative tasks. For anything else, we should give preference to contract work, which allows us to obtain certain expertise such as accounting or legal services when needed, or support community projects such as software development without creating a permanent position. Even there, we should proceed cautiously. The board's decision to assist some of our overburdened volunteers with paid work is quite recent, and we need to take the time to evaluate whether it was a good move. There are also considerations such as the appropriate level of pay across national boundaries which we haven't properly addressed.
Perhaps even more important than the number of paid positions are the guardrails we put in place. To me, that includes forgoing paid management or leadership positions. We should preferably recruit long-term community members who understand and respect the project's principles, and have proven that they can work without close supervision and openly cooperate with volunteers rather than through employee-only back channels. We should pay close attention to transparency, guarantee community involvement in the definition of new paid positions, and nurture loyalty to the project as a whole rather than the foundation or board. If you've read the Hiring Framework adopted by the board, you'll be familiar with these and similar points. This is not a coincidence, as I was deeply involved with the creation of that document, and consider it a good foundation for our next steps.
Your views on the use and control of AI (Artificial Intelligence) edit systems?
I believe AI technology is best used as an additional data source or tool set that is made available to human mappers to help them improve their productivity. It can be valuable in that role, and I find it encouraging that the popular systems seem to be trending in this direction.
All imports and large-scale editing efforts, including automated and organized edits, can cause damage, and adding AI to the mix does not fundamentally change this. If they are done poorly, they can harm the community and drive people away from the project, which far outweighs the benefits of broader data coverage. Pitfalls to avoid include low data quality, non-standard tagging, and poor communications with local volunteer mappers. For individual volunteers, it is easy to feel helpless and frustrated when faced with a group making unwelcome changes. Because of this power imbalance and the other inherent dangers, the OSMF has a set of rules to govern such activities, and a body (the Data Working Group, DWG) to enforce them. I believe the DWG is doing a good job, but if a situation cannot be resolved, the OSMF board is available for appeals as well.
One additional risk with AI is becoming dependent on any particular provider of AI technology. The best antidote is diversification, much like we now have access to a large number of imagery providers. It also helps to choose open source solutions: Our popular editors are open source, which ensures that we cannot lose access to them even if an entity funding the editor's development were to lose interest in OSM. I would prefer for our AI editing tools to be open source as well.
Finally, the question asks about my personal experiences: While I have checked them out to satisfy my curiosity, I do not use AI-based tools in my own mapping at the moment. They don't really help much with the kinds of mapping I enjoy most.
What is your opinion of the proposal for a software dispute resolution panel?
As a board member, I have contributed to the proposal for the software dispute resolution panel, and I support giving this new institution an opportunity to prove itself. Specifically, it does stand a good chance of resolving some of the issues concerning the iD editor, and as explained in a previous question, I believe it makes sense for the OSMF and community to take an interest in iD development in particular.
If the panel proves worthwhile, it should preferably become self-sustaining by handing the responsibility over to one or several working groups. I would like to avoid relying on a group of people hand-picked by the board for this role. The board traditionally refrains from taking on tasks that could be advanced by working groups and the increased centralization this would imply. Our working groups, such as the DWG, bring relevant expertise to the table, and have built up a high level of trust over the years.
Should we do anything about EU database rights?
Our goal should be a solution that preserves the existing level of protection within the EU if this is practically feasible. It is unlikely that we would be able to move the OSMF itself in the short term (although I would not rule it out indefinitely), but we are currently investigating more plausible approaches such as the creation of an EU subsidiary.
OpenStreetMap has come a long way from its humble beginnings. Billions of people are using our maps, either directly or as part of another product. Our data powers some of the world's most popular websites and apps, serves the needs of hikers and cyclists, helps optimize logistics, propels video games to the next level of realism, guides cars navigating parking lots on autopilot, offers accessibility for people with disabilities, adorns postcards, wall clocks or clothing, and unlocks countless creative and unexpected possibilities beyond. OSM is everywhere. By producing a truly global dataset, our project offers enormous transformative potential, and a fertile ground for innovation where even small organizations and individuals have unhindered access to the world's best source of map data.
We've clearly done a lot of things right! If we want to see this incredible success story continue, I believe it is essential to preserve the values that have made it possible. To me, this includes our appreciation for local knowledge and the open-ended data model which caters to a diverse set of use cases as well as our spirit of do-ocracy that allows everyone to chip in and empowers people with valuable skills and great ideas to make a difference. And most of all, it has to include our amazing community of contributors who must not be – as is the case with all too many platforms relying on user-generated content – a mere source of free labour, but the group which controls the direction of OSM.
As our project grows further, attracts the attention of a growing number of commercial actors, and has to deal with a changing social and technological landscape, keeping the OSM spirit alive requires us to respond to these new challenges. So even though I'm happy to have been able to implement many of the ideas outlined in my 2018 manifesto, the overall goal of strengthening the influence of volunteers in the OSMF is sure to keep me busy for two more years.
That's not all I want to get done, though, as there are a few corners of the project where it's hard to shake the feeling that forward movement has stalled. Notably, whereas the wider ecosystem is healthy and vibrant, some of our core services have not kept up: The data model hasn't seen any major new features in over a decade. Some of our communication platforms are aging and, at times, insufficiently moderated, which contributes to an unfortunate trend for subcommunities to scatter across various proprietary channels. The openstreetmap.org front page is selling OSM's capabilities short by showcasing only a limited subset of our data, and there are missed opportunities to present our project in a better light. As the entity operating these core services, the OSMF has a responsibility to meet the growing needs of the OpenStreetMap project.
For a more thorough exploration of my ideas for the board, and some of the specific challenges that I expect we will have to deal with as a community, I invite you to also check out my answers to the official questions! This manifesto isn't meant to duplicate everything I've written elsewhere, but as a spotlight on some of my core priorities.