Who Really Owns OpenStreetMap? A Personal Concern About Big Tech in OSM

General talk
21

That’s quite an interesting topic!

I think I would frame a bit differently your concerns. The first thing is to differentiate: are you talking about governance or are you talking about data?

There were some attempts, from big companies, to control the foundation (employees trying to run for the board, for example). It didn’t work out well, and in the big picture, they 100% gave up on that (see the millions they paid and are paying to create/maintain Overture), at least on the short term. Regarding the data, while I dislike ODbL, I don’t see how one can “own” the data, due to the licensing terms. The data is “free”, so anyone can use that.

What I see is, while you don’t need to pay to use the data, you kinda have to pay to truly use it. OSM does not offer an easy way to professionally use the data, so you have to rely on external services for that. Also, OSM doesn’t offer “clean” snapshots of data, so, again, you have to rely on external services for that (think of that free Facebook service that I forgot the name, which is now inside Overture). Overture itself is an attempt for that: provide an easy and reliable way to ingest data.

So, answering your questions in a broad way, I would say: relax, they don’t really care about us anymore. They are happy with the free labor (they pay some people too to fix/improve the data), and that’s it.

The question is: after knowing all of this, are YOU happy with that? If yes, just keep mapping, don’t question yourself that much, and you’ll be fine.

(regarding your 4 specific questions, I am more concerned about some people inside OSM community than from big tech. They are the ones shaping OSM priorities and profiting from that, and you didn’t even notice that.)

1 Like
22

That said, big tech and it’s literal tens of billions of dollars really should be paying the OSMF and our volunteers for our contributions. To the best of my knowledge, nobody’s doing this. Previously, Osmand, probably the player least capable of doing so, did pay mappers in Bitcoin (though I’d strongly prefer something that, you know, actually spends in the real world).

And I don’t mean providing their own paid mappers (the attempted hostile takeover a few years ago honestly makes me believe that corporate directed mapping, paid or unpaid, should be expressly prohibited), I mean paying existing unpaid volunteers a living wage in their current country.

2 Likes
23

Such really unprecedented support is vanishingly unlikely. Some more realistic goal would be more viable.

2 Likes
24

The Situationist doctrine of recuperation might be relevant here:

the process by which politically radical ideas and images are twisted, co-opted, absorbed, defused, incorporated, annexed or commodified within media culture and bourgeois society, and thus become interpreted through a neutralized, innocuous or more socially conventional perspective. More broadly, it may refer to the cultural appropriation of any subversive symbols or ideas by mainstream culture.

Which makes me think that the challenge for OSM is to double down on the things that make us unique. Map places for people, not just for cars. Build tools for indie developers and cartographers, not just for megacorps. Maybe (as per @Kugelbaum’s point above) we need to do better at spreading the “maps for people” message beyond the developed world, rather than just relying on Big Tech contractors to fill in the car- and delivery-focused basics everywhere else.

14 Likes
25

We do, but the overall process is too fragile. Continuity, development, quality, tooling, unity, decision making, documentation, it’s all too… too… well, let’s stick with fragile.

2 Likes
26

That doesn’t mean that we shouldn’t still be asking for it, loudly and frequently. Corporations need to pay into the society that allows them to exist.

3 Likes
27

And constantly hamstrung by people refusing to acknowledge, much less move forward, on removing car-forward stumbling blocks. Like the folks insisting on car-route refs on ways, and only tagging car lanes…

2 Likes
28

redefining lanes tag is offtopic also in this thread

why you try to fit it literally everywhere?

4 Likes
29

Hello, Stumbling Block! Have you met Wall? Wall is more productive to talk to.

30

As long as information useful to cyclists and pedestrians can be recorded and used, I’m a happy bunny.

If we were starting from scratch we might do a whole bunch of things differently, but if people had spent the late 2000s arguing about how to count bike lanes rather than “just mapping things” we probably wouldn’t be here now.

10 Likes
31

I’m not faulting the “just map things” approach, I’m faulting OSM’s tendency to consider in-the-moment decisions sacrosanct and immutable instead of allowing things to evolve equitably.

1 Like
32

There are plenty of ways that OSM tagging has changed since say 2014 (not least the number of tags per object - I bet that is much more now than it was). Where there has been change it has been to allow the mapping of a new concept (highway=busway for example) or because the previous tagging didn’t allow other data to be mapped (highway=ford on ways).

There have been relatively few successful mass retaggings to a different tag with the same meaning (like waterway=riverbank - acceptance of that one was eased because data consumers mostly handled the “new tags” already).

In the case of lane counts on roads, I think it’s fair to say that there hasn’t exactly been a groundswell of opinion supporting your point of view, and even if what you suggest has excellent arguments in favour** without support it isn’t going anywhere.

** I don’t have a view either way on it in terms of “which would have been best in 2007”, but given that the same data is stored, think that we should not change the way we count lanes if there is no actual benefit to doing so.

3 Likes
33

Except data consumers already support inclusive lane tagging (because using lane counts as checksums is obvious) and generally choke on situations where the number of lanes tagged for turns, access or destination exceeds the lane count. The status quo isn’t supported and creates off-by-n errors.

34

Obv. i’d love if AppAmaGooFaceSoft gave “OSM” millions, but be realistic, they’d never just “give” that much money, without expecting to “set priorities for OSM”.

7 Likes
35

That doesn’t stop us from shaming them as ungrateful tightwads…

2 Likes
36

Well, the question is whether we want to get some donations from them (though less than paying good full-time wages for each of us), show our superiority and distaste or achieve something else.

37

C: Get compensated fairly for the time and effort they’re profiting very generously off of. It’s the least they could do.

38

Wouldn’t mind being paid €1000 a month to contribute to OSM

2 Likes
39

Well, I would not mind being paid €1000 a month to map in OSM, getting veto power over EU law, moon base and replacing highway=unclassified highway=road by more intuitive tag names while we are at it.

The problem that gap between this and anything likely to happen is absurdly vast. And trying to get this is (A) waste of time (B) may make harder to achieve actually possible things.

8 Likes
40

And like it or not, that using OSM data costs less than running your own mapping department is part of the appeal.

5 Likes