The OSMF – changes during the past year and what they mean for the coming years – part 2


In the first part of this blog post i summarized the most important developments in the OpenStreetMap Foundation (OSMF) during the past year from my perspective. In this second part i will – based on the observations from the past year and recent trends being visible – give a lookout on how the OSMF could develop in the coming years.

After my previous piece on the OSMF and the perspective for the upcoming general meeting, Severin Menard has published an interesting and likewise critical take on the current situation of the OSMF. I agree with most of what he wrote and his culturally somewhat different perspective on things is very valuable. Definitely recommended for everyone to read.

Corporate takeover – it has already happened

There is one thing however i disagree on with him – the assessment of a corporate takeover as a risk of the future. During the past weeks there have also been some half baked initiatives started (last minute before the general meeting as usual) from within the OSMF board towards resolutions meant to protect the OSMF against external takeover. My view is that these measures and the focus on a defensive strategy against an external attack aimed at controlling the OSMF are meanwhile setting the wrong priorities because the corporate takeover has essentially already happened behind the scenes without that being clearly visible to the outside.

Large corporate OSM data users are not really that interested in staging a coup in the OSMF and run the OSMF themselves at this time. That would be expensive to do and bear a large spectrum of fairly big risks and also it would – if successfully executed – immediately lead to a struggle for control between the major corporate players. The main goal of corporate actors with the OSMF is and has been for some time to prevent meaningful regulation of corporate activity in OSM and of OSM data use based on the OdbL. If that goal is secured, a secondary goal would be for the OSMF to serve as a shared neutral intermediary platform between the different corporations to steer independent volunteer activity in the OSM community in the corporate data users’ collective interests.

The parts of the OSMF board not affiliated with corporate or organized interests seem to have, during the past years, developed the idea that these corporate interests can be negotiated with and that the future of OpenStreetMap lies in compromising between the organized and corporate interests and the core ideas and values of the project. This quite clearly is an illusion – expecting a big corporation like Facebook to make compromises with an insignificant player like the OSMF is at best naive.

If the goal to prevent meaningful regulation of corporate activity through the OSMF has been accomplished permanently, corporations have no reason to oppose effective takeover prevention of the OSMF – because that would help protecting their position in the OSMF against third parties gaining influence. And effective prevention of meaningful regulation does not even depend on there being a pro-corporate majority among the OSMF members because the corporations have other significant channels of influence in the OSMF now (through their financial constributions, through their participation in the working groups and through lobbying of the board on non-public channels).

We will in the near future have a fairly good test case for how robust the ability of corporate interests in the OSMF is in preventing meaningful regulation in form of the attribution guideline that has been worked on during the past years. There are essentially three scenarios of what could happen:

  • The OSMF board decides to adopt a guideline roughly based on the corporate wishlist from the LWG with some minor adjustments for the optics, to maintain the impression that it is not directly what corporate lobbyists have written. That seems the most likely scenario at the moment but it bears the strong risk that the craft mapper community will openly oppose this interpretation of the ODbL which would fundamentally endanger the OSMF’s position in the OSM community.
  • The OSMF board adopts a guideline reflecting the community consensus reading of the ODbL (likely similar to what i drafted) that unconditionally requires attribution that practically makes the user aware of the origin of the data. That would be strongly against the corporate interests and corporations would certainly do everything within their power to prevent that (including withdrawing funds which would leave the OSMF in financial peril because the strongly increased costs make it depend on regular corporate contributions).
  • A decision on the matter is avoided by dragging out the process indefinitely. Although not ideal, because it would be a kind of unstable situation, this would be acceptable for the corporations because it would maintain the status quo of the OSMF not becoming active against data users with insufficient attribution. It would also avoid an open break with the OSM community, although there would be likely increasing pressure from the OSM community on the OSMF to get active in cases of insufficient attribution by the OSMF’s corporate financial contributors.

Some will likely reject my idea that the corporate takeover of the OSMF has already happened. They will argue that if that was true, corporations would push much more aggressively for their interests. I don’t think that is the case though. As explained, the primary interest large corporations have in the OSMF is not positively accomplishing something, it is preventing things negative for their interests from happening. Accomplishing that is much more valuable for them than anything they could proactively try to push for in the OSMF.

What we will certainly see in the coming years is corporations trying to consolidate their influence on the OSMF, in particular by more and more corporate employees being encouraged to volunteer and being paid for work on the OSMF in the working groups, on the board, in committees and other ways. This will happen rather quickly in most of the working groups probably, supported by the move of the OSMF to position itself more like a corporate actor which, as i have explained before, is likely going to have a negative effect on motivating volunteers without career interests in OSM for the OSMF. My estimation is that in 1-2 years a solid majority of the people engaged actively in the OSMF in one way or the other is either employed in some form in an OSM related job or otherwise has a carreer interest at least partly motivating their involvement in the OSMF. Most of them will be employed by corporate OSM data users or organizations around OSM like HOT. Currently in the OSMF board for example we have already at least three members to whom this applies, two corporate employees (Mikel and Paul), one small business employee (Rory).

Centralization of the OSMF

The other big upcoming trend in the OSMF i can observe is an increased centralization of power towards the board. The OSM community overall is highly decentralized and the OSMF has always been kind of an abnormality within that with its hierarchical structure. But traditionally in the OSMF most of the work has been done in the working groups – including development of policies – and the working groups had a high degree of independence, starting from being created by grassroot initiative from within the community, to allowing also non-OSMF members to participate and to by convention allowing board members to contribute, but not allowing them to have a lead role. We more recently, however, see a trend towards more and more policy being either actively influenced by the board or being developed by the board itself from the start. Early manifestations of this trend were the already mentioned Crimea decision (where the board overruled standing policy developed by the working groups) and the organized editing policy (where the board flat out rejected the first draft of the DWG and demanded a more lenient policy). This year we saw a large number of cases where the board created internal policy, often presenting the results as a done deal without having an open discussion – like in case of the diversity statement – further emphasizing this trend.

In addition, as i have discussed in the first part, we saw during the past year the establishment of several committees (a concept that previously did not exist in the OSMF) – put together by and under direct control of the board. And for this year’s general meeting we have an AoA change proposed that allows the establishment of committees consisting of board members and OSMF members and to delegate any powers of the OSMF board to these. In other words: It would allow the OSMF board to recruit volunteers or paid staff (paid by either the OSMF or by third parties) and delegate any kind of function of the board to them. Obviously, such committees would be under immediate and absolute control of the board, people could become members of such committees only by appointment by the board and the board could dissolve or remove powers from such a committee at any time. But, in contrast to the working groups – which don’t have any formal powers and depend for any meaningful decisions on approval by the board – the commitees could be equipped with any of the formal powers and rights the board has within the OSMF.

The effect the establishment of such committees would have is a massive shift in power within the OSMF from the working groups towards the board. Currently the board is essentially limited in what they can do by their numbers. Board members are not able to delegate their formal powers to others. Work requiring more hands can only be done by the working groups, which have a high degree of independence. If the mentioned resolution passes, the board would essentially no more depend in any way on the working groups of the OSMF – they could assign any tasks so far in the remit of the working groups to the committees they appoint and control without any restrictions.

We can already see right now that the board is increasingly trying to take direct influence on what the working groups do and how they work. In public communication they have expressed an interest in reshaping the remit of the CWG and to reactivate the EWG (which – since the EWG is completely inactive right now outside Google summer of code management – amounts to nothing less than bootstrapping a new working group, something the current board at the beginning of the current term still considered inappropriate).

An interesting test case for the tendency of the board to seek direct control over what happens in the OSMF is the establishment of the planned software dispute resolution panel. The DWG has expressed interest in this task but it seems likely that the board will prefer to directly control the composition of this panel.

What is not visible is any indication that this trend in centralization of power in the OSMF is balanced in any way by more independent oversight over decisions and processes within the OSMF. The idea, for example, that the local chapters could gain some kind of meaningful power in the OSMF’s structure and processes, or that the OSMF could share control over key elements of the OSM infrastructure (database rights and the contributor database) with the local chapters, is not in sight. Given how keen the current board seems to be to gain more immediate control over everything within the OSMF, it seems unlikely that the board will be willing to relinquish any power voluntarily any time soon.

Decreasing diversity and brain drain

Another trend that has already been visible for quite a few years is the increasing difficulty for the OSMF to attract competent and committed volunteers. As i have pointed out in context of the plans of the OSMF of paying people for work, this will likely have a significant negative effect on volunteer motivation.

Now, if you see this problem only as a numbers game, it is not unlikely that it is possible to fill this deficit with paid people (paid either by the OSMF or by external interests) or by people which view volunteering in the OSMF as a career builder. As indicated above, this development is already in progress. There are a number of effects this is going to have:

  • Parts of the OSMF (working groups, committees) will become increasingly dominated by culturally narrow circles of people with shared interests (like their careers or shared interests of outside organizations they are affiliated with) – interests which are distinct from the funadamental goals and values of OpenStreetMap. This process is going to be self emphasizing because once this happens the working group or committee in question becomes increasingly unattractive to anyone outside this narrow spectrum who does not share the interests of the group – even if that group of people considers themselves to be open and welcoming to others.
  • It will probably be in particular the deeply committed OSM community members who have many years of experience in the project for whom volunteering in the OSMF will become increasingly unattractive and who will likely seek to contribute in functions outside the OSMF. That means there is likely to be a quite significant brain drain of competency and experience.
  • The de facto goals of the OSMF will increasingly drift away from the goals and values of the OpenStreetMap project to the special interests and culture specific values of those who happen to be dominating the organization. From within the OSMF and its communicative echo chamber it will potentially not be so visible – the prevailing opinion on the OSMF board often already seems to be that that OSMF’s and OpenStreetMap’s goals are neccesarily identical, in an L’état, c’est moi kind of way.

Seeking influence on OpenStreetMap

As a result of what i wrote in the previous sections it seems likely, and there are already indications for it, that the OSMF is going to increasingly try to take influence on the OpenStreetMap project itself, getting in conflict with the basic premise of the OSMF of supporting the project but not controlling it. The fields in which this is likely going to happen are in particular:

  • Community communication channels: One prevailing narrative from within the OSMF more recently has been that “fragmentation of communication” in the OSM community is a big problem that needs to be addressed. That terminology is in itself interesting by the way – the same thing, when considered positively, is called diversity, when deemed negatively it is framed fragmentation. But that just as a side note. We can expect there to be strong pushes from within the OSMF to try taking tighter control over communication channels hosted by the OSMF by imposing behavior regulation. If these attempts will be successful will need to be seen. What can be said with near certainty is that if that happens it would have the opposite effect of what it claims to want, it would more strongly fragment the communication within the OSM community by essentially squeezing out those who do not want to subject themselves up-front to a narrow culture specific communication rule set in their communication and who would move to communication channels outside the control of the OSMF.
  • The OpenStreetMap website: Traditionally the OSMF has no say in the design of the OpenStreetMap website – it is developed in the classic OSM tradition of do-ocracy and consensus. But there have been increasing voices from within the OSMF expressing the desire to less prominently feature the map as the symbol and the main instrument of inter-cultural cooperation in the OSM community and re-design it more like a corporate website and more proactively communicating the OSMFs interests to the visitor. We will likely see a move in that direction in the coming year – if that is going to be successful is not sure in my eyes though.
  • Mapping and tagging: The OSMF board has last year already made a move that could lead towards taking an influence on mapping and tagging in OpenStreetmap, something that is in principle strinctly forbidden by the mission statement of the OSMF, through the plans for a software dispute resolution panel. The primary purpose of the panel is resolving conflicts w.r.t. development decisions of the iD editor which are primarily conflicts about tagging presets and validation rules. Deciding on such conflicts (which is what such a panel would need to do) would inevitably amount to making tagging decisions. The OSMF of course has no means to enforce such decisions currently beyond their influence on the iD development, but still, the potential of attempts in that direction is definitely there. As i have explained above, a secondary interest the financiers of the OSMF have is steering the craft mapper community into a direction beneficial for their data uses. Members of the current OSMF board have already expressed that they deem usefulness of the geodata collected in OSM as the primary goal of the project – in contrast with the traditional goals and values of the project which put in foreground community cohesion. And if you assume usefulness of the data to mean usefulness for the economically important corporate data users’, interests of the corporations and goals of the OSMF board seem to align here. Still, in contrast to the previous two points where clear trends and already visible, on this matter i would not make a clear prediction.


Now this outlook onto the next years is obviously rather bleak and, indeed, my view of the near future of the OSMF is fairly dark. For OpenStreetMap in general i see this, however, as a valuable chance to emancipate itself from the OSMF a bit more – which, no matter where the OSMF is steering, would be a healthy thing. And i also want to point out a few chances i see for the OSMF in the near future. As i will explain, the likeliness of these actually happening is quite small – yet i find it important to show that the OSMF is not inevitably doomed, but that what happens still depends on the people in positions of power making responsible decisions and there is still a possibility for developing into a much more positive direction. Two examples:

  • With the Brexit and the possibility that a move of the OSMF from the UK to the EU is advisable, there is a real chance in the context of such a move to restructure the OSMF from the highly hierarchical and centralized form essentially required by British company law, into a federated organizational structure with checks and balances and meaningful subsidarity rules, as it would be appropriate for an organization within the highly decentralized OSM community. This is, unfortunately, highly unlikely to happen even if a move of the OSMF takes place, because it would essentially require the OSMF board to decide to give away much of its power to more federated authorities in a new organizational structure. But the possibility is there – there are no meaningful principal hurdles against implementing such a change beyond the general difficulty of moving the organization in the first place.
  • The OSMF has the chance, due to the strongly increased public visibility in the last years, to gather the necessary means (and i mean this independent of contibutions of larger corporate financiers with either explicit or implicit strings attached) to return to and to concentrate on positioning itself as a neutral infrastructure provider for the OSM community. With that i mean to – like a state providing roads and other infrastructure to its citizens – provide neutral infrastructure for everyone to use without discrimination. This idea is not the same as desiring a very small OSMF. Infrastructure can be extensive and expensive. If the OSMF would want to offer each local community around the world the infrastructure to design and run their own real time updated map for example, the costs of this would be fairly massive but at the same time it could be highly beneficial for supporting cultural diversity in the OSM community. The key here would be neutrality and non-discrimination. Much of what the OSMF currently does in new money spending is not – the already cited idea to financially support people whose work we know and enjoy is fundamentally incompatible with this idea. Unfortunately, like in the previous point, this vision would require restraint from the OSMF board to resist any urge to govern and actively steer OpernStreetMap in a direction they deem desirable – something i do not see being likely to happen with the current board.

As i wrote in part 1 of this post, i present my prediction for the direction in which the OSMF is headed here to be proven either right or wrong by what will actually happen. And i would be happy to be proven wrong – either by what the future will actually bring or beforehand by convincing arguments brought up by those who see a different development coming. In other words: I openly challenge anyone who disagrees with my analysis to present and argue for their own predictions in public.

One thing the OSMF board seems to have lost over the past year is the willingness to expose their vision and their plans and ideas to an open discussion and try to convince the OSM community of the merits of their ideas in open discourse. Probably at least in parts because with the money the OSMF has, it seems easier to just buy the implementation of your plans instead of engaging with and convincing the community to support you.

And i call to everyone in the OSM community to not accept this. Whether you have the feeling the OSMF is going in the right direction or not, whether you agree with my analysis here or disagree: You should require people in the OSMF to present arguments and reasoning that convince you of the merit and the solidity of their plans and their actions and require them to present a meaningful vision and expectations for the success and outcome of their actions which can be tested against the actual future development just like my predictions. What is best for OpenStreetMap is not the median of all interests articulated, weighed with the amount of money behind them. What is best for the project can only be decided through arguments and reason.


  1. > there have also been some half baked initiatives started (last minute before the general meeting as usual)

    Nope, I’ve discussed those AoA changes inside the the board for 9+ months, and in general for more than a year.

    re: Attribution Guidelines, There is no substantial difference between your draft and the “2020v2”, if anything the latter is more clear, and has _less_ wiggle room than your draft.

    • Thanks for your comment, Rory,

      the Community attribution advice you probably refer to is not in any way comparable to any of the OSMF guideline drafts – it has the primary intention of formulating the community consensus on interpreting the license while none of the OSMF drafts even attempts this.

      And it is crystal clear on that – when writing the text that was the primary goal. The core of that interpretation is:

      […] the criterion for a valid attribution is if it effectively makes the user aware that OpenStreetMap data licensed under the ODbL is used […]

      There is no wiggle room here. What the text does not attempt and what you are probably missing is providing the data user with a guarantee that meeting certain formal criteria makes a data use universally acceptable in terms of attribution and frees the data user from any further obligation. That is because there seems to be no community consensus on that, the prevailing opinion seems to be that if a certain form of attribution is acceptable depends on if it actually effectively makes the user aware. And – as i write in the concluding remarks – that the honest effort to do so counts more than abiding by certain formal requirements.

      If the board wants to discuss the community attribution advice or wants me to explain the text and its intention and meaning in more detail i would be happy to do so.

  2. I would like to make three small points:

    (1) “The corporate takeover has already happened”. I think that this is not the case. You try to draw a fine line between a “coup” which you claim has not happened, and the not-specifically named corporate takeover that *has* happened. You rightly say that once a “coup” were to happen, corporations would quickly start to jostle for control among themselves. I venture that this would also be a problem with the lesser form of corporate takeover that you claim has already happened. But what I do see – and this was also a point you wanted to make and I overlooked it – is that, especially with more money being spent, there’s a danger of what we in German call “vorauseilender Gehorsam”, where board members themselves will start to think “but what will our corporate sponsors say”, and “maybe we should invite them for a talk”. That isn’t a takeover per se, and from an organisational/legal point of view it could present itself as the prudent way to govern, but still it would lead to an undesirable rise of corporate influence. (In the cruellest of cases this could lead to the board doing or not doing something in an attempt to please corporate donors and the corporate donors not even wanting that.) But, I see a light on the horizon in the form of big charitable foundations like those bankrolling Wikipedia or HOT; persuading them to part with a tiny but of their cash for OSM would greatly reduce the role of corporate money in OSM. It would still come with strings attached, but not the greasy corporate kind.

    (2) “My estimation is that in 1-2 years a solid majority of the people engaged actively in the OSMF in one way or the other is either employed in some form in an OSM related job or otherwise has a career interest at least partly motivating their involvement in the OSMF.” – You have worded this so widely that indeed a lot of OSM-related work falls under it. If someone saying “I hope that some day I can get a job in Geo that’s what draws me to OSM” is enough, then the corporate takeover has likely happened 10 years ago already.

    (3) You cliam the board is attempting to take direct influence on the working groups. I can’t speak for other WGs, but as a member of DWG I cannot complain about board interference since the last board election. In fact, only recently a corporate mapper has requested “guidance” from the board about their activities, an opportunity that previous boards would almost certainly have grabbed for a little limelight, and the current board has immediately bounced that issue to DWG as it was part of our remit. I’m well aware that this can change at any time as board majorities fluctuate, and you rightly point out the troublesome birth of the organised editing guidelines, but I’m just saying, at the moment we don’t have any reason to complain.

    • Hello Frederik, thanks for your comment.

      Regarding (1) – i don’t think we are that far apart in our assessment here. My definition of the corporate takeover having already happened (and i probably should have made this clearer in the text) is that the OSMF board is incapable of making decisions that are in substance against the interests of the corporate data users, in particular w.r.t. regulations of corporate activity of any kind. Of course incapability inherently cannot be proven, only disproved.

      Regarding (2) – yes, of course there are always borderline cases but the vast majority of cases are very clear. I am pretty sure if we’d do an inventory of OSMF volunteers right now there are less than ten percent, likely less than five percent where there is any doubt if they are to be classified as hobby volunteers or as people with a professional or career interest related to OSM. It’s like in mapping, you have to do a classification and draw a line somewhere to formulate meaningful statements. My point here is not tied to an exact definition, my point is that i see a clear trend that the hobbyists in the OSMF are in retreat and people with professional and career interests are on the rise. And not in a subtle way, massively – so that the trend will be visible independent on what your exact method of measurement is.

      Regarding (3) – the DWG is highly atypical as a working group in the OSMF, it is the only WG with a clear function in OpenStreetMap outside the OSMF and as such has earned a reputation within the OSM community that is highly independent of that of the OSMF as a whole. And it is also the only one which has developed a truly robust independence from the board. Therefore i fully believe you if you say that there is no visible attempt at taking influence at the moment. But that does not mean that the board would not take the chance to either weaken or take influence on the DWG if they had the chance to.

  3. Pingback: weeklyOSM 541 | weekly – semanario – hebdo – 週刊 – týdeník – Wochennotiz – 주간 – tygodnik

Leave a Reply

Required fields are marked *.

By submitting your comment you agree to the privacy policy and agree to the information you provide (except for the email address) to be published on this blog.