Personal take of a longtime OSM contributor: I don't think it's an issue and welcome any improvements from any directions. I also don't care if megacorps use the data to make jillions of dollars and never contribute. But I know others feel differently.
If someone manages to take over OSM and put a restrictive license on the data, we'll just fork from the last open version and keep on doing what we do best: we map!
I work in, love, and use almost exclusively open source software and I don't agree with your description. We wouldn't be anywhere if it was only passion projects. For me, I like open source because it avoids wasted effort: everyone works to push the state of the world forward, instead of hoarding their own work for their own profits. It's simply the best way to make software. But that doesn't require doing things only because it's fun and meaningful. Doing work on open source software because you can get paid to do it is a perfectly valid reason to work on open source software.
I don't think he's saying that his take should exclusively be the reason to work os OSS, but rather he thinks that the freedom to use OSS should be the same for everyone.
You both state the major reasons why people do it.
That's a fair point. Perhaps my statement paints too broad a brush. I get paid to work on open source software, so I agree it's not only worth doing because it's fun and meaningful.
I guess I feel lucky hitting the sweet spot of enjoying the work I get to do on open source and be paid for it at the same time.
Bless you for being robust. This is the spirit that will save humanity from regulatory spider webs, and produce some fantastic things long into the future.
Lobbyists are a thing, as are barriers to entry. Of course they love regulations that help them. They sometimes even write those regulations themselves.
My notion of regulations isn't about quantity, but about quality. The lower the quality, over time, the higher the quantity. Thus forming a sort of spider-web-like structure, except spider webs are often of greater quality than legislation.
The article is a blurry overworded FUD, because fear get more clicks (yes, I clicked too), and Bloomberg probably pays/evaluates by length. There's no clear issue mentioned with corp contributions, just what-if-something-maybe stuff.
As far as licences go, though, the terms and conditions regulating usage of OpenStreetMap now fill not three documents, but three SECTIONS on the OpenStreetMap Foundation's wiki. However liberal these licences may be, for the money it would cost me to hire a lawyer to read through them, I could just buy access to a commercial service.
Sorry to hear it, but it seems that plenty of other applications felt the lawyers were worth it. I'm curious what exactly you believe the risk to be, except I don't think this comment is really made in good faith
I find the license very confusing - does any dataset generated from the database have to be released freely to anyone who asks? Or is it only a customer that purchases the dataset? Where is the distinction between a product of the db and the db, which are subject to different terms? For anything which isn’t displaying a standard base map it is confusing.
> I also don't care if megacorps use the data to make jillions of dollars and never contribute
I can see you've received lots of praise for your commitment to open source, your passion about mapping, your contribution to the world. Regardless, the above statement just seems sad and depressing, the acknowledgement and resignation.
That's a weird and telling use of the word resignation, that says a lot about your mindset.
Either you didn't mean that word, or you projected into parents' comment something that you believe but that he didn't say nor seem to believe himself.
If I knew that hundreds, or thousands, of hours of my personal volunteer time and effort, were going to be exploited by some MegaCorp to make billions of dollars, I'd be pretty pissed, and discouraged to continue, even if I loved the project and mission.
If that makes me a selfish person (or less commendable than the OP), so be it.
How does them making, or losing, billions of $ affect the pleasure you get?
The farmers ploughs the fields and harvests the crops, and earns his reward. Others will take that crop and make billions more in distribution, retail sales, as ingredients in other foods and so on.
Each works for his reward, and if the satisfaction you get is not enough then you need to seek a space elsewhere in the value chain.
If it is enough then don't worry about how others extract their value - it doesn't impact on you. The Baker is as worth of his bread as the Farmer is of the wheat.
>If I knew that hundreds, or thousands, of hours of my personal volunteer time and effort, were going to be exploited by some MegaCorp to make billions of dollars,
I'd never be worried about dollars at all after that.
It's arguing that corporations are contributing data to OSM, but that this somehow... changes the proportion of what kinds of things OSM covers? That traditional users have added "hyperlocal features, such as a neighborhood bench or informal walking path" while coporations are adding "roads, buildings, and all other points of interest" and "a particular emphasis on improving road data".
How is this anything but good? As long as data is correct, the more data the better -- proportions are meaningless. Buildings and roads aren't taking away from benches and paths. And I'm pretty sure buildings and roads are benefiting all users anyways, no matter who contributes them.
I'm utterly baffled by what logic the author imagines this is a bad thing. The data is open, so it's not like corporations are winding up with exclusive ownership of anything.
One of the typical ways this manifests itself in reality is a dedicated local mapper adding or correcting a detail where they know that some common data sources are either wrong or out of date. At some point later a corporate mapper the other side of the planet working from imagery will come along and change it back to agree with the imagery they have, despite it being 4 years old. Local mappers get exhausted from the effort it takes to keep their patch from being broken and disenfranchised with their edits being drowned out by the noise of the corporate machine.
A devils advocate answer of course, I'm personally on the fence on the issue.
That's the kind of thing that happens when someone is paid pennies per edit and given insufficient resources to guide those edits, which really only happens when someone points a hose of money at a contractor.
I've long been a fan of a mandatory time field! Now to get Stack Overflow to implement it too :)
That, my friends, would be the first step for my dream to finally map historical data, like in the OHM, directly in the OSM without anyone feeling bothered by it.
That hasn’t been seen in practice though. Apple has been making changes from surveying with their own vehicles, not out of date imagery. Amazon makes updates from delivery drivers reporting issues making deliveries.
This specific thing has been seen in practice, it was fairly regular occurrence back when I contributed to OSM and a source of lots of similar complaints as this article at the time. In fact, often the argument was framed as automated vs human editors as some non-corporate large-scale data imports had gone badly wrong too.
Even at that time though, it seemed like the corporate support was, on balance, a positive thing. The main worry at that time (for me anyway) was some corp getting smart and pitching themselves as a middleman for other big corps. Luckily the contenders all got slurped up by big corps in a feast of mutually assured destruction, and since no one could trust their future to be built on a competitor's playground, they seem to have all defaulted to OSM.
Similar to Linux, having a diverse set of medium sized corps on your team (and maybe a few really big ones that see you as a complement, not a competitor) seems to be a winning strategy.
One of the typical ways this manifests itself in reality is a dedicated local mapper adding or correcting a detail where they know that some common data sources are either wrong or out of date. At some point later a corporate mapper the other side of the planet working from imagery will come along and change it back to agree with the imagery they have, despite it being 4 years old
It's happened to me. Not just on maps, but on Wikipedia.
I think the concern is big companies might get into governing positions and make decisions about the projects future... For example, they could decide to divert some efforts into a commercially licensed 'extended' dataset, or they could sign restrictive deals to get hold of 3d laser scan or satellite/aerial imagery that isn't so easy for volunteers to collect.
That would be a legitimate concern but the article focuses so much on stuff like "look how much edits Apple made!!" that it's really hard to tell if it was a concern for the author, at all; or if there are others (e.g. it spends about as much time talking about the governance concern as it talks about diversity in the governance body; so it's unclear if author is worried that companies might take over, or is just unpleased with the governance in general).
Same as the parent poster, I'm baffled. This article makes no sense to me. If it had a point to made, it buried it in thousands of words of blabbering.
If we look very well at it we see Apple made the most contributions to map Africa. Who's going to complain about that? Google, maybe? I guess african roads don't suit the google maps cars.
It's also a curious business decision... Apple doesn't make much revenue in Africa, nor have many users there. Why direct all their mapping efforts there?
Africa is the largest growing continent and just like China has been an extremely important market for many companies, so too will many developing African countries. Apple would love to be the market leader there and any headstart they can get will pay dividends in the future.
But Apple products are generally extremely expensive ( usually at least multiples of the cheapest equivalents), and that kind of thing doesn't sell well in underdeveloped developing nations, which cover most of Africa. Just look at median income and GDP PPP in China, and any African country. The mere fact that China is actively (kindof) neocolonising multiple different African countries should be telling as well.
No-one said that Apple directs all (or even most, or even just a significant proportion) of their mapping efforts there. Should they ignore a continent?
Yes exactly, a American country with terrible roads.
How can Google map roads in the Americas with their terrible roads?
We don't make distinctions between any African countries anymore so I think it's only fair we start treating the other continents the same way.
(I mean of parts the US are close enough to third world countries anyways, 11 year olds dying in mobile homes because corrupt leaders hobbled basic utilities and all)
"American" implies it's in the USA, like if someone is described as "American", they're from the USA. European, Asian, African, Australian, North American, and South American all refer to the continent, Antarctican seems to be a similar word (without much use, unsurprisingly), but I don't think there is one for the whole Americas in English, because "American" isn't it.
There's no other word for people from the United States because of the weird naming of the country.
Canada - Canadians, Mexico - Mexicans, Brazil - Brazilians.
United States - ......United Statesians?
That's like people from the DRC calling themselves 'Democratic Republic of the Congolesians.' Nope, they just shortened it to Congolese. The United States of America just did the same thing. 'Americans.'
They all use it, and for the most part so do their neighbors.
Until some clever soul figures out a better way for the bologna of the North American sandwich to refer to it's citizens, the largest collection of English speakers on the planet will probably keep on trucking with the same boring term.
It's incredible how dense people can be when their world view is so narrow.
The point isn't there should be a word for it, the point is Africa is not a damn country. Such obvious facetiousness but thick skulls can only fixate on "wHuT amErIcAnS Are thE US"
South Africa has as much in common Sudan as the US has with Haiti
There are plenty of countries on the continent than can (and have been) mapped by Google Maps.
Saying that Africa is somehow inhospitable to Google Maps is exactly like acting like North America is inhospitable to Google Maps because Haiti has bad roads.
Hell I'd take driving on the roads in that Haiti picture over some in places like "Chiraq", probably less chance of a bad outcome in a Google Maps car.
In that article the Indian pop is from the 2001 census at 1,028 M. Current population is 1,388 M, keeping the same percentage of English speakers would place it at +305M speakers, although doing the same for USA also place it at +310M.
That being said I think the table is wrong because the numbers don't add up to the total and also don't match the source. Based on that census India should have only 12.18% of English speakers, not 22.18%, so you should be correct in any cases.
I'm not American, or from the Americas, I'm just letting you know the meaning native English speakers will take from the word. Call non-Americans American if you want, but don't be surprised when you're misunderstood.
> You should learn the meaning of the word "facetious".
I know it well.
> And I'm not sure what makes you think I'm not a native speaker of English.
You apparently don't know the definition of "American". I have never seen a native speaker make a mistake like that.
> Apparently you don't even understand blatant sarcasm, are you one?
I understand sarcasm just fine. Calling a Haitian road "American" is not sarcasm. Yes, English is my first and only language.
You said a picture was from the USA, and I doubted this, so I went and spent time chasing down the actual source. That's not facetiousness or sarcasm, that's either an innocent error, or misleading and antisocial. I interpreted it as innocent error, and attempted to politely correct your knowledge and explain, and now you reply with this argumentative comment, implying you deliberately attempted to mislead and waste our time. I won't reply further.
You're just too dense to understand what was said, you shouldn't have replied in the first place.
Person implies somehow Africa, the continent, is somehow unsuitable for Google maps vehicles (despite the fact Google has mapped hundreds of tens of thousands of miles in multiple countries in the continent)
This comes from treating all of Africa like it's one giant country with the same infrastructure everywhere.
I facetiously (a word you do NOT understand apparently) show a picture of Haiti, a North American country and say, since their roads are terrible all North American roads are terrible, and refer to an entire continent as if it were a single country that includes the US.
From here your subpar English skills fail you, to the point I'm forced to use a word somewhat similar to facetious but not quite as accurate (sarcasm) because the proper word is above your understanding.
Sometimes being imprecise allows a person with a challenged vocabulary such as yourself to catch up, but of course the Dunning-Kruger effect is in full force here, so you use it as an assertation that you're not the one failing to understand basic vocabulary.
It's a shame English is your only language and yet mastery of it escapes you.
Thanks. I guess I just ask, is there any basis for that fear though?
The OpenStreetMap Foundation's governance [1] isn't tied to who contributes data. Is there any evidence they are being "corrupted" in any way at all by corporations adding data?
And like any open project, if its governance were captured by "bad guys", users can always fork it and "vote with their feet" since the data's open.
If OSM really was being subject to a hostile board takeover by corporations then obviously that would be hugely newsworthy and concerning. But it's like the article is trying to stoke fears of this while presenting zero actual facts to support it.
Based on previous large company involvement in 'open' projects, yes. It usually starts out fluffy and friendly and just 'wanting to help'. The large company (or companies) take on an increasingly large role eventually leading to taking on and then over governance... then it's often game over as far as the original intent of the project. By the time they are in the governance roles, the dissenters have typically already been marginalized and/or are gone.
This is a very old strategy. Companies (especially the large, name-brand ones) aren't getting involved because they believe in the mission/objectives of OSM but rather that they see a strategic advantage in their involvement. They dip their toes in, get entrenched and then make their move. It's a lot faster and less expensive for them than trying to replicate what the project has produced.
I'm very interested. Can you point to specific examples?
And did the original contributors fork the project to a new successful initiative? Or if not, why not?
I'm also curious why companies feel the need for this -- if the data is all open in the first place, why do the companies even need to bother? Why don't they just use the data directly?
Without specific examples it's hard to understand.
Specifically related to the story, Apple has a rather troubled history as an Open Source 'contributor'. They tend to not contribute unless they need to and then prefer to run the show. They basically steamrolled KHTML with a fork of their own code a.k.a. WebKit (which they effectively control development of exclusively). Imagine them doing something like that to OSM... which they will if it ever suits their purposes. See most other Open Source projects Apple has been involved in... it's not pretty. So yeah, Apple playing a significant role in OSM should have alarm bells ringing in the heads of other contributors.
That's not to say corporate involvement/sponsorship isn't important: it's vital to many important projects. The issue is when companies start trying to influence the projects beyond their technical needs and start trying to wrap their business model around them (a good indicator of that seems to be their involvement in governance) that the trouble starts. I think that's the concern being expressed about OSM.
Is there really precedent for companies taking over open-source upstreams and turning them proprietary? I expect the corporate users appreciate the price & freedom at least as much as everyone else.
Not sure what "open-source upstream" means, but in software, there is the complaint that companies like Google can add a lot of complexity to software projects, which serves Google, but not the classic home user of Linux. That can make it effectively their own corporate project and drown out more casual developers.
Ironically OSM is the opposite. The hobbyist mappers do bus routes and admin boundaries and mountain trails and all manner of complex stuff. The corporate mappers are really just interested in the most basic element: drivable roads.
That much is true - direction and feature set. The accusation that seems weird is that companies would be motivated to make it proprietary just because they use it. At most, it seems like you’d get a niche for “enterprise” distributions with support contracts on the same stuff.
Oracle is the most (in)famous via various acquisitions. But IBM/RedHat/Canonical/Microsoft are a few other large players that have been trying with varying degrees of success recently related to Linux. Corporations appreciate ever increasing EPS and decreasing costs... nothing more.
For me this discussion is new, so I don't have good answers to these questions.
One thing that happened when Pokemon Go got popular is that people (users) were conducting vandalism, in the sense that many pokemon figures were added to the main map of OSM. People have to work really hard to remove these, and they might burn out on this. There is already a lot of vandalism, like people adding their friends drinking shed, or whatever they think of as funny.
Another thing that already happens with volunteers, is people thinking their working area is "their area", while no such concept exists. This can already give a lot of friction among volunteers. I can imagine if billions in money are at stake, big companies with a massive crowd of emplyees might want to really grab those working areas and drown out volunteers. If that gets a big problem, volunteers that were active for years might drop out.
Oh, one other thing. Many volunteers are maintaining and guarding hiking routes and cycling routes. These get damaged easily (I just did that yesterday myself) and these volunteers sometimes carry a big burden. The big companies don't have an interest in these routes "by the people, for the people", there is no money in that. If they get more often damaged, it will not be regarded as funny by these important long-lasting volunteers.
Is this virtual or real damage? As in, the trail route is imperfect and misleading about the actual trail, or the trail is revealed via the map and others come and physical use damages the trail. The second reminds me of land owners sneakily putting "no trespassing" on land that is legally publicly accessible (I think this happened in Sausalito about a year ago)
It's often as simple as mis-tagging. I do a lot of mapping of mountain bike trails (think narrow, hiking-type trails which are also open to bikes) and these are designated as paths (highway=path), open to cycles (bicycle=yes or bicycle=designated) and some mountain biking-specific tags.
What'll often happen is a well-meaning person will come by, often using an editor that doesn't differentiate between types of bicycling, and change the route to be tagged highway=cycleway. This is the tag for things like paved bike paths, such as urban cycling infrastructure.
The result is now these mountain bike trails, narrow/rocky/tight/steep stuff, shows up in tools that offer bike commuters directions. Just... not great.
So many of us keep an eye on the stuff in our area to see that the tags stay as they should.
There might be many situations, but I was mostly talking about virtual vandalism. When you edit points, lines or routes in the wrong way, they get damaged. The volunteers guarding these need to fix them themselves, which can be a burden and they can get emotional already. You can imagine how that goes when more damage gets done where the fixing is left for those volunteers.
I don't know much about land-owners doing vandalism, often people try to communicate, sometimes this works out okay.
Is this really such a problem? I guess this is my first time hearing of it, but I believe OSM allows you to mark areas where public wifi (or private wifi, in the case of businesses that provide wifi to customers), is available. Wifi access is not a tangible, physical thing (even though of course the physical layer is), but OSM provides a framework to map it because it's tied to a location, and the information might be useful to some people.
I am not sure what you are replying to, so I assume there is some miscommunication.
I don't know anything about mapping public Wifi access, like in shopping centres.
What I meant is that already some people feel like their city and a 30km radius around it is "their area" and they want to control what and how things get mapped there. It can lead to heated and emotional discussions, we are people after all :)
If a big company with lots of human resources would do this, there is hardly any fighting back.
I think the issue was that Pokemon Go used certain types of map elements as their Pokemon capture points, e.g., Landmarks, or something. People weren't adding "Pokemon capture points", they were adding non-existent Landmarks so that they could be imported to the game for their own advantage.
OSM contributor here: agreed with you, not sure what they're on about. All the sources are also old and known, with the newest references to the latest State of the Map conference. It's a long article with a lot of random historic fun facts, but I don't quite get the point.
> When it launched in the mid-2000s, most spatial information was owned by governments, and was difficult or impossible to access.
It bothers me that this sentence is used as if the facts behind it are a logical arrangement of consequences. The fact that people are prevented from accessing data that they funded, and by all rights own, is an absurdity.
I appreciate that OSM exists, but in a perfect world, it shouldn't have to.
"Difficult to access" here may have meant going to your national/state mapping bureau (or however it's called) and physically buying a printed copy of the maps. It's not necessarily that they "prevented people from accessing data that they funded", or at least it was not around here, just that getting maps required physical movement which I guess these days qualifies as "difficult to access".
I (and several others) used to trace these maps (by hand, using OziExplorer) then offer them for download in Garmin/MapDekode
format. OSM didn't even exist at the point.
It was possible to obtain a digital copy, but you needed special equipment.
Most government data was only available on 9-track tapes. So few media outlets had the necessary equipment that reporters' groups would publish step-by-step information about how to get it done, and with what hardware.
Yes, but digital maps are old enough that my elementary school had a networked GIS package to show us maps in the computer lab in the early 90s, and it's much older than that.
That hasn't always been true. USGS topo maps, for example, were originally drawn and plotted on paper and had to be digitized. At one point, I think they had a way to request a digital scan, but there was a processing fee.
Now almost all of that data available online for free from their website, and they have ways to access it in bulk and via API.
I think the bigger problem nowadays is accessing GIS data from different cities and local governments. Many states and cities put the data online, but each one organizes it differently, possibly in different formats, with different methods for accessing it.
Some of that government data existed in the form of physical maps, and the data was difficult to access because it had never been digitized, even government-internally. No one was prevented from digitizing it themselves, and in fact a lot of early OSM data came from enthusiasts digitizing old US Army maps because, indeed, they were public domain.
Then you have cases like the Ordnance Survey in the UK where apparently a lot of the data is neither freely available to the general public, nor paid for by the general public.
The OS are making a lot more data freely available as part of the Geospatial commission push to make geospatial data more widely available to drive innovation. I wouldn't be surprised if this is all a slow response to external activities like OSM and Google
I remember reading that when all OS data cost money half of their income came from other government departments, so once you add dealing with all that licensing there's not much extra cost to just fund it out of general revenue.
> a lot of early OSM data came from enthusiasts digitizing old US Army maps because, indeed, they were public domain
But that's the rub, we the taxpayers have paid for all the map creation, even the new high resolution ones with far greater accuracy, yet we're stuck with these older, lower resolution and likely out-of-date maps. Even the new maps should be public domain.
If you're referring to maps produced by the US government, they are in the public domain. Early on it sounds like there wasn't an easy way to transfer "raw" map data around (hence the tracing), but it's pretty easy now to access official maps in full resolution:
Those linked above typically available through a standard API (Tile Map Service), as well as ESRI's moderately well documented REST API for accessing shapefiles and features. My one complaint about this data is how the USFS manages their data. In that case it seems each service region defines how data should be made available, so you have a pile of different methods for accessing things like road/trail features, or points of interest (in many cases, it comes down to "fetch this pile of ZIP files").
In the Netherlands we have access to, and are permitted to use and derive from, government map sources such as the cadastral maps and the various municipal data layers (trees, benches, natural features such as waterways, streets).
This is great, and we can go the extra mile on OpenStreetMap because of it, but these sources are not a replacement for OpenStreetMap; the goals are different and the data presented on it is different. On OSM points-of-interest are important. Things like shops, attractions, schools, museums, etc. Government maps don't have these. Municipal maps may have streets drawn in as areas accurate to the centimetre, but that is not the same as the graph of routable ways that OSM has (not just for cars, but for bicycles and pedestrians as well).
OpenStreetMap mappers can make great use of government sources where permitted, but a good digital map is much more.
Well not exactly. E.g. in the Netherlands we have open building and address data. It took the community, with many volunteers, like a year to import that into OpenStreetMap.
Compiling open data into a comprehensive map without duplicate data or rough edges is a lot of work. I agree that all governments can make this a huge lot easier by just publishing what they already made with public funding, but that doesn't mean OSM would suddenly cease to have a right to exist. It would just be able to improve its quality even further.
In a perfect world, rather, commercial maps shouldn't have to exist.
I agree with your point, although I wouldn't go so far as to say OSM shouldn't have to exist. There is still a lot of value in a global, collaborative mapping project that can be updated by anyone, on the ground, within seconds of a change. Official government maps can take much longer to show new buildings, road changes, etc.
Conflation is a huge accessibility concern and it’s not possible to mandate cross-government without some unified international government body. Even within the US, spatial data conflation is a major technical and political problem at the federal level, since the data is mostly collected at the state and municipal level, then rolled up into a Frankenstein dataset. The truly global nature of OSM is unique and the geographic diversity of coverage is unrivaled even among commercial sources, much less governmental.
I don't know about other countries but here in Sweden we have Lantmäteriet ("The land measurement agency") and they have an interactive map based on their data that is free to use. https://minkarta.lantmateriet.se/
I am very pleased that corporations are helping us with the map. It can sometimes be annoying when somebody who isn't from the area messes some complicated intersection based on old satellite imagery. It is awesome when a corporation adds large numbers of features that I never would have added. More contributions is a net positive.
The only concern I have is that corporations may seek to prevent OSMF from offering competing map services. Mapbox might, for example, seek to hamstring the openstreetmap.org tile server in order to push people to their commercial offering. This is largely theoretical but is a large concern of mine. As long as core technologies are Free/Open Source, we can always pack up and fork if it becomes problematic.
The status quo is (roughly) to resist adding capacity to the osm.org servers anyway. Not so much to push people to commercial offerings, more to avoid growing the resources required to operate the foundation and site.
The Crux of the article I feel: Open Street Maps (OSM) is free and the rest largely aren’t. Conversely the cost of things like Google Maps (as a service not as a consumer in this context) has risen dramatically so naturally big companies are turning to the free offering more and more and the worry is they will have outsized impact in the project and become the dominant body and their priorities will be pushed forward over others
And not a whiff that they are contributing back in substantial ways on the whole either
Seems like all giant corporations are one way or another not respecting the spirit of the mission of the organization and often aren’t receptive to community feedback in any substantial way
Edit: maybe leeches (that was my previous statement) is harsh, however I stand by my assertion that the spirit of the project isn’t being taken into consideration in this instance. Whenever FaceBook, Apple etc can make it proprietary they will rather than share data and compete on things like user experience. They could do much better in this regard. I don’t believe personally that it’s the developers at that company per se that are the issue this is an issue with industry politics and open source policies etc.
But they are contributing back, that's the point of the article right?
So much so the article is more about who controls the organisation.
I personally see this mostly as a good thing. More data, better data, and everyone gets value out of it: users and corporations. Surely it's a sign this project is doing things right.
However I think it's right that the article points at the power structure of this organisation, I think now more than ever it's important it remains independent.
Sometimes they are not contributing. There have been complaints about contractors employed by the big firms adding non-existent roads, tagging roads wrongly, etc. The concern is that because the big firms are so big and now have power over OSM governance, they can simply ignore all those community members who are noticing their persistent mistakes and asking them to stop.
This seems to imply that the corporate members are inserting these mistakes on purpose. I am not convinced this is the case, as I do not see how they could benefit from creating bad maps.
It is not that they are inserting the mistakes on purpose, but rather the contractors actually doing the edits are not responding to changeset comments saying "Hey, you've been persistently making a tagging error with regard to X, could you please stop?" The big corporate members are so big and powerful, they don't care if their contractors are sometimes violating OSM etiquette and failing to work together with other members of the community.
If you ban the account of one employee of a contractor, that employee can just make a new account, or the contractor company can give that employee another account to edit from.
Why assume that? Maybe there's a bug in their internal software, and it's cheaper to create a program that checks if you're banned and automatically creates a new account, than it is to fix their internal problem.
It is presumed that sometimes they don’t respond to changeset comments because they aren’t proficient in English (or the local language of whatever country they are mapping remotely).
Hobbyists who keep getting their changes reverted and their account banned will eventually give up, especially if they're not actively trying to harm the project.
But a contractor who's paid for their work, even if it's not very good? They'll just keep going so they don't lose their job.
Hobbyists driven by passion tend to evaluate their work on quality. Contractors mostly evaluate on quantity, especially if their pay depends on such quantity.
It is like open source debate. Contributing back could either mean hiring/giving massive sponsorships to the core contributors, or it could mean giving out scraps. A couple meaningless lines of code here and there and the corporate sponsor gets to pretend they have fulfilled their social responsibility. The bulk of the labour is still born by those who toiled for years without material reward.
the complaint seems to be less about anything practical, and more about how you used to get a warm fuzzy feeling from contributing your data to a scrappy upstart, and now that OSM has achieved such success and is being used by so many companies, the warm fuzzy feeling has gone away.
not sure that's really a problem that needs solving.
> And not a whiff that they are contributing back in substantial ways on the whole either
How have you achieved this conclusion when the entire article is about how corporations are contributing back, and the fear that those contributions might cause the devaluation of volunteer contributions? There’s even half a dozen visualisations to demonstrate the extent of their contributions.
I should clarify: I don’t mean just material code contributions but other contributions too, like monetary, data, being receptive to the community about contributions offered and such. Respecting the ethos is more in line with what I’m talking about
It’s a tussle, because I personally came away that the ethos of the community isn’t being respected and the fear is they’ll overrun the mission of the organization.
It’s complicated and nuanced to be sure., which is why I amended my original comment
In my experience, it’s not even the ethos of the whole community, it’s the ethos of a few old guard European dudes. Ask nearly any mapper from say Southeast Asia what they think of corporate contributions and I think you get a different answer.
>Conversely the cost of things like Google Maps (as a service not as a consumer in this context) has risen dramatically so naturally big companies are turning to the free offering more and more and the worry is they will have outsized impact in the project and become the dominant body and their priorities will be pushed forward over others
I don't know if this has changed with newer versions of the google maps API, but years ago when I had the choice of using google maps or OSM, the biggest motivator in my decision, when OSM was still a lot less complete, was to do with Google's licensing.
There was a clause that implied they'd have some rights to reproduce and use our data and that was totally unacceptable for our use case. OSM was completely free to use without worries that google was somehow going to be able to use our data in some way.
Trying to piece together what the actual concerns are:
* Mapping is not purely factual, some things can be mapped in different ways and corporations might map things that conflict with how normal people see them. Maybe an example would be deleting a commonly used path through cooperate headquarters that was made for staff/visitors.
* Too much data in niche areas like driveways overwork volunteers and steers attention away from citizen areas of work
* Governance model, fears corporations will get too much control through sponsorships
I see these as valid issues but I think they can be worked through. For me it has been extremely nice to have a free and open data source for projects and the OSM data model is really easy to use I think. I feel the benefits outweigh the problems but I am only a user looking from the outside.
As a long-time OSM contributor, 1 is the only one that bothers me.
We've had a bit of an issue in the UK with Amazon Logistics mappers blatting away some of the nuance in rights-of-way mapping (which is quite a complex subject in UK law) as they map driveways for their own use.
That's not a massive problem in itself - OSM has reasonably good communication mechanisms (particularly changeset comments), and there's a clear Organised Edit Policy which most of the big guys follow. It's just a challenge because of scale: it's hard for part-time, unpaid volunteers to spend hours chasing after the countless Amazon contractors who are sitting editing OSM all day. I know we've lost a few individual contributors because of this, though more in the US than in Europe.
But this is mostly growing pains. With goodwill from all sides I don't see any reason it won't be resolved.
OSM has a rather straightforward way of mapping things: if it's there, we map it. If the USA claims to own Poland, then we do nothing. If the USA actually occupies parts of Poland, then we change the tags to reflect that reality. Not to indicate that they own it, perhaps, but at least to indicate that it's stably under their control. Conflicts (I'm a moderator in an OSM chat, it can get very heated) mostly arise from people who don't like reality and want it to be shown differently. But their issue is with the renderer, e.g. OsmAnd, for showing it as belonging to what they perceive to be really theirs and unfairly occupied. But that doesn't change what the underlying data should show: reality. I'm having a hard time thinking of what would create a true conflict without clear solution.
Re-reading this in the morning, just to be clear, this will actually steer routing engines away and is not just something for humans to read. Routing engines take the access tag into consideration and it can also be specific to certain kinds of traffic (e.g. no cycling allowed here, no trucks, no pedestrians...).
Of companies mentioned in the article, the asymmetric contributions to OSM that bother me the most are AllTrails and Strava - 2 companies that heavily rely on trail data sourced from OSM.
AllTrails does close to zilch to help put trails into the maps, even though the majority of trails are user-generated content.
Strava has let "Slide", their one project that could help put trails on the map, die an ungracious death.
If you build a whole business model on top of free data, it may be worth considering improving that data.
Strava worked with Mapbox when they were switching over. I don't know what exactly the deal was, but Mapbox employees added a bunch of missing stuff to OSM based on Strava data analysis.
but I do find this whole discussion strange, because it's rather obviously the same problem that open-source _anything_ has: you're giving something away for free, and it's very hard if not impossible to control how that data gets used or whether the primary benefactors contribute back anything to the source.
Ironically, the only known model of forcing people/corporations to contribute something back based on their usage of something is called "market-based pricing".
Unless someone is proposing that maps are somehow ethically or systematically different, than, say, Linux, this conversation feels rather unspecific and pointless.
There's a similar problem with websites which map rock climbing routes. Some make their data open to some extent, but not in a way that can be meaningfully contributed back to OSM. Even though they all use OSM data (usually via Mapbox), to generate their maps, and allow users to draw and annotate layers on top of it.
I'm very interested in working towards an open tooling and open data ecosystem for rock climbing information, but I don't know where to get started on finding others to build it with.
Does strava actually use the trail data from OSM? as far as i can tell, the routes on strava are entirely contributed by strava users and don't come from OSM in any way.
The two maps follow the same paths obviously, but at least in my region the routes and segments on strava don't ever seem to start or stop at the trail intersections in the OSM maps and often cross unmapped and unofficial connectors that don't appear on OSM. The extent of their reliance on OSM trail data seems to be that they use MapBox tile images and those tiles sometimes have OSM trail names marked on them.
Strava's map are based on OpenStreetMap, there is attribution in the bottom right corner of every map (not sure where in the mobile app, I'm not an active user).
The parent comment probably talks about giving Strava user uploaded data/tracks back to the OSM community so they improve the map (or add new paths themselves). In a way they already do https://wiki.openstreetmap.org/wiki/Strava#Data_Permission_-... Allowing tracing is more than other companies offer and gladly accepted. It regularly leads to new unmapped paths discovered, sometimes on private ground (military areas) that are otherwise inaccessable to an OSM mapper.
yeah, that's what i was saying about the tiles - their tiles come from mapbox, which means they include OSM data and therefore must include the attribution. but the claim was that they're heavily reliant on OSM's trail data, and I just haven't seen any evidence of that.
I work in geospatial stuff and we use OSM data. There's things I'd contribute back but the community is notoriously prickly. You make a couple of mistakes not made out of malice and they'll assume you're corpos out to fuck everyone because that's what evil corpos do.
The downside is high (you get a bad reputation as being an evil corpo) and the upside is limited. So, for the moment, we just hold the corrections on our side and overlay them on top.
If you're an actual corp you don't care about that, but being labeled an evil corpo will ruin things if you want to cooperate. And we're a small startup.
I'll revisit it some time but we're being cautious about it.
I don’t want to reinforce your feeling (I can imagine exactly the pain you’re talking about) but isn’t there a legal requirement for map corrections to be published? They don’t have to go back into OSM, but they do have to be made available because of the share—a-like concept of the license. At least, I thought so.
You can add Map Notes which aren't seen as threatening to the extent some apps that use OSM as their base map let their users directly send them in https://wiki.openstreetmap.org/wiki/Notes
This isn't welcome behavior (I think there's not significant technical limitation on it though).
The things lots of people in the OSM community want to see are willingness to cooperate and willingness to communicate. Making throwaway accounts obviously doesn't accomplish those things.
(there's value into letting people quietly make contributions; reducing community hassles associated with systematic changes isn't really where that value lies)
Well it was an honest question, not a suggestion. I suppose there are good reasons for that policy, just seems a shame to lose out on good honest updates because people are afraid to post without anonymity.
I've read the whole article and I don't get what its trying to say. Corpos are not contributing enough? They are contributing too much and thats changing the culture?
There's a fairly large number of long time contributors to OpenStreetMap that only want driveways if they are added with deep passion (I'm like ⅔ serious).
There is a real issue where it doesn't make sense for a small volunteer organization to manage a massive amount of data that is not of interest to a broader set of users. It's not a huge issue, because the narrow interests mostly understand the basic issue, there's just not universal agreement about what is broadly useful or not.
I had the same impression. Amazon moved into our area and someone from their logistics has been contributing a lot to OSM mostly documenting service roads. They have the resources to really add to level of detail of OSM. I see it as a net positive.
I thought that was the point, everyone improves the part they care most about and in the end we have something very comprehensive. You even get people who care a lot about filling gaps just for the sake of completeness. So long as nobody corrupts the big picture it should be fine.
"Such devaluation could lead to what he called “digital gentrification,” in which the very attributes of OSM that drew its earlier users are degraded by its newfound fame."
A look at much of the internet and the declining role of academic guiding principles wrt technology and its governance tells you where this might be headed..
Hope OSM somehow can find a golden middle way that incorporates "grassroots"/cute data like benches as well as large-scale and tedious to maintain data such as roads or road signs. Don't let 'em eat you.
This article seems to take the absurd position that these companies contributing to OSM are making it worse. One can only imagine the article written in the alternate universe where Apple et.all used the data but didn’t contribute their improvements back - any chance at all it would be a positive one?
Am I reading this correctly that some volunteers are upset that their edits represent a smaller proportion of total edits because big companies are adding lots of data? But in absolute numbers the volunteer edits are still the same?
If that’s the case, is that essentially asking for a map with less data so their own contribution can be a bigger fraction of the total data?
I thought they were talking about wikimapia.org, still upset that Google hasn't offered them a free API key or something like that, that project is awesome. Among other things it was one of the best ways to follow the Syrian Civil War, at the beginning, at least.
The wikimapia.org isn't bigger is really a shame. That site is like a treasure.
To be fair, how interesting it can be totally depends on the community of your city. The place I used to live seem to have one or two people that are very passionate, almost all the buildings are marked and often have interesting tidbits and trivia attached to them. I used to spend hours reading it. The city I currently live, on the contrary, is relatively dry in that regard. Still useful, but not something I can binge-read on.
On the technical side though, I feel like they're always lacking even before the whole GMaps API thing. It has so many bugs (and site is slow), to a point that viewing or editing become annoying from time to time.
I was very passionate about Wikimapia as an actual wiki for everyday people to add to in the first few years. There was a lot of graffiti, but it was also building into a great resource for local interest and history. Then the cost concerns picked up and deletion wars began. The emphasis was on removing all places and content not considered "significant" and current. Real estate and business became the primary focus and the wiki aspect fell away. I loved how that site started.
It seems rude to mention Strava in conjunction with withholding data in this way as I'm fairly sure they have helped to contribute back anonymized gps traces to the community for tracing of biking and hiking routes.
Similar to BSD licence, this is beyond their legal requirement and is some kind of token of good faith in what seems a mutually beneficial relationship. I'm sure they could do more and it's an open question on how to get those that benefit most to contribute in similar proportion but I'd think there must be better targets to call out.
I understand the potential nightmare scenarios that the author describes, because the author is worried about the sheer amount of "editing capabilities" of a company compared to a small community of OSM editors.
But doesn't that automatically imply that the current way of verifying OSM data is flawed and needs a better architecture?
Setting aside my personal opinions about feasibility... I think that this is actually a valid use case for a proof-of-map based system that removes the statistical power of a company with too many (maybe unintended) potentially malicious editors.
Corporations have a broad set of methods to defend themselves from bad actors. They also have money to set them up, or influence own direction.
It is an interesting question how open project should defend against such actions. Without money and without a law protecting them.
As an enthusiast of open source movement especially, it is worrying that such projects are easy targets for wrongdoing. Examples of such actions could could fill a book or two. This is happening for years.
Someone you know might have mapped that! Local data usually comes from the locals. If you enjoy it, one easy way to also contribute is with the StreetComplete app, which just asks you questions like "is this street lit" (my girlfriend actually asked me for that data once, if I knew how to figure out which roads are lit, so she could go running after dark and plan a good distance without going into dark areas).
This comment breaks quite a few of the HN guidelines. These, for example:
"Don't be snarky."
"Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith."
"Eschew flamebait. Don't introduce flamewar topics unless you have something genuinely new to say. Avoid unrelated controversies and generic tangents."
"Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."
Can open source wiki/user contributed style tools implement a type of captcha to protect against big corps' influence? I see major entities regularly fiddling with wikipedia articles to insert bias. Why not turn Google's Recaptcha back on itself?
The article isn't about Wikipedia articles but about OpenStreetMap. Unlike articles it's much harder to add bias to data points, e.g. a street or position of a house, that can be verified.
I thought this article was about control over OSM, not sure why it's so difficult to make parallels between this and wikipedia. Selectively adding accurate data points can very much bias a map, e.g. all you see is McDonalds and never a local restaurant
If McDonalds were to add all its restaurants to OSM what would be the downside? They are not removing other restaurants from the map. I fail to see how adding accurate data points would hurt OSM even if it is done "selectively".
Cherrypicking causes biases [0]. A highly non-random distribution of data points is the issue here. No one has to remove other restaurants, it's just that there could be 10x more effort put in to add detailed McDonald data points.
I don't see how that follows at all, it's very easy to make things disappear (a CEO's 3rd vacation home perhaps) and getting ground truth on a whole planet is a hard problem.
If someone manages to take over OSM and put a restrictive license on the data, we'll just fork from the last open version and keep on doing what we do best: we map!