7 ms·

I tried making an app that would leverage OSM late last year and was disappointed at the amount of false or stale information in there. I went into the experie

I tried making an app that would leverage OSM late last year and was disappointed at the amount of false or stale information in there. I went into the experience thinking it was the wikipedia of maps. But it's not, wikipedia has much better content control than this. My theory is that it's too easy for random people to enter data into OSM. It should require some sort of validation of the contributor. And each contribution should have a discussion, just like wikipedia talk pages. I know that sounds like a lot of talk pages but I believe some of those contributions require community discussion to improve.

rathel 6y ago
Examples, please. There is bad info here and there, but it's minority (if you want 100% accuracy it's delusional, even Google Maps has a ton of crap, starting from skylights that AI recognized as buildings [3] to businesses you can't even tell they exist). Also, for a long time now each changeset can be discussed. [1] I have a hunch you may be from a country where the community has not took up too much [2]. These tend to have more dubious data. [1] https://blog.openstreetmap.org/2014/11/02/introducing-changeset-discussions/ https://blog.openstreetmap.org/2014/11/02/introducing-change... [2] https://osmstats.neis-one.org/?item=countries https://osmstats.neis-one.org/?item=countries [3] https://goo.gl/maps/knpT8eRR4YhB6Sm86 https://goo.gl/maps/knpT8eRR4YhB6Sm86 - look at the shopping mall in satellite and map view
- ucosty 6y ago
  Not OP, but I'm in the UK and find the OSM data to be not all that useful. The streets themselves seem mostly accurate, but there is almost no building data. If you had to rely on OSM data to get around, your best bet would be to live and work in northeast London, within the circular.
  - Doctor_Fegg 6y ago
    YMMV really. OSM in the UK has orders of magnitude more cycling and walking data than Google, for example, but fewer buildings. I don't generally find I need buildings in my day-to-day mapping needs but everyone's different I guess.
    - bpye 6y ago
      I have found this in the U.K., Canada, Japan and Taiwan - OpenStreetMaps is way better as a pedestrian or a cyclist in my experience, at least in relatively built up areas.
  - the_pwner224 6y ago
    Its the same in many parts of the USA. For example in a college town with 50k students / 50k permanent population, streets are complete with sidewalks and bike trails etc., but good luck trying to find nearby restaurants by looking around the map, like you can do with google maps. Even the buildings which are marked with their name and type usually lack operating hours, phone number, website url and other sorts of info that GMaps has. But the situation is much better - close to GMaps level - in big cities here.
    - matkoniecz 6y ago
      > lack operating hours If you have an Android phone - I want to plug StreetComplete app here. I never mapped opening hours before I started using it. It asks simple questions, there is no requirement to learn anything OSM specific. Registration at OSM website is probably the biggest hurdle. (disclaimer: I am one of contributors to that project)
  - smichel17 6y ago
    A great thing about OSM is that you can improve the data, specifically in your local area so that at the very least it will no longer have spotty coverage of the places you frequent. I use https://github.com/westnordost/StreetComplete https://github.com/westnordost/StreetComplete for doing this so I don't have to make the investment to learn the technical details of tags.
    - drstewart 6y ago
      "You have to create the map to use the map" isn't as compelling of a feature for a mapping service as you think it is
      - nelgaard 6y ago
        OP talked about improving OSM using StreetComplete. So this is about adding missing speed limits, opening hours, roof shapes, recycling containers, tactile pavement at bus stops, etc. Hardly "creating the map". When I travel to a place where OSM could need improvement I make sure that at least the hotel I will be staying at, nearby restaurants, bars and museums, etc are correctly tagged with phone numbers, opening hours etc. It is not much work and it gives me an opportunity to decide what I want to do before I leave. When I get there I then have everything ready offline in OsmAnd. In a way I just update OSM instead of adding my own bookmarks/favorites.
      - smichel17 6y ago
        nelgaard got the gist of it. I've never encountered an area where OSM didn't have the roads and buildings already mapped. I have encountered areas where it didn't have the speed limits and house numbers of those structures (and, once or twice, places lacking a road name, but that's pretty rare). But, my personal perspective was a little different. It's more like: I want to use OSM, and get others to use OSM instead of proprietary mapping solutions like Google's, for moral reasons. In that context, being able to make sure my friends can always get to my house using OSM is a great boon. I realize I didn't share that perspective in my original post, which is probably the source of the confusion here.
  - 7952 6y ago
    The great thing about OSM is the currency. New developments will be added more qucikly. But for completeness and consistency Ordnance Survey data will just be better. And they do have a free building dataset, and even pre-generated vector tiles.
    - gnufx 6y ago
      Even the OS data -- at least what I can see in Aurora -- might not be as complete as where someone has taken an interest in a new development, such as one near here. The OpenData building shapes are a good start, but they appear to have been auto-generated, and usually need correcting from up-to-date satellite imagery, if not a survey.
  - gnufx 6y ago
    Whether buildings are mapped, and in what detail, just depends on the area; detailed mapping definitely isn't confined to London. The main utility of mapping buildings for navigation might be to tag them with postcodes, even if that's not ideal. Roads are typically at least as accurate as OS OpenData where I've looked.
- c0nsumer 6y ago
  Roads in rural areas are a HUGE problem. Particular in remote / forested places where logging roads were used in TIGER data. For example, places like this: https://www.openstreetmap.org/#map=14/46.6397/-87.7149 https://www.openstreetmap.org/#map=14/46.6397/-87.7149 There's a number of us who use this kind of data for building back country cycling routes, and while it's great guidance, there's both the problem with roads being listed that are no longer passable (nor often even visible at ground level), or other routes available that don't exist. Of course, this gets back to the way much of the data is crowd sourced... And we're working on that. :)
  - gerbal 6y ago
    Anecdotally, in Northern Wisconsin, GMaps is only slightly better than OSM on abandoned logging roads. There are a lot of private gated or abandoned roads and logging tracks which Google Maps will use for route planning.
    - c0nsumer 6y ago
      That's kinda how it is here. The best thing seems to be to bounce between Google Maps and OSM and an aerial view to get a sense of what's up. For an interesting example, check out this place: Google: https://www.google.com/maps/@46.4133003,-87.5248774,5345m/data=!3m1!1e3 https://www.google.com/maps/@46.4133003,-87.5248774,5345m/da... OSM: https://www.openstreetmap.org/#map=14/46.4161/-87.5406 https://www.openstreetmap.org/#map=14/46.4161/-87.5406 Neither has the actual current lake boundaries (as used as a retention pond), Google has roads where the lake is while OSM doesn't, etc.
      - maxerickson 6y ago
        3dep can be a huge help in figuring out how prominent tree covered roads are. https://elevation.nationalmap.gov/arcgis/rest/services/3DEPElevation/ImageServer https://elevation.nationalmap.gov/arcgis/rest/services/3DEPE... It's listed as a Theme Overlay at https://viewer.nationalmap.gov/services/ https://viewer.nationalmap.gov/services/
- INTPenis 6y ago
  I wanted to use node data on shops, like if they take cash or not for example. Noticed there was already a lot of shops with data in there. But none of it was accurate. Just one example is that an Apple store in Sweden was set to accept cash. Just for fun I called them up and asked if they take cash and they laughed at me. It seemed to me that someone had automated input for businesses without actually verifying their data.
zelphirkalt 6y ago
It sounds like a good plan, however, there is the question of how you want to validate the map information. Someone adds a street. What do you want to do to validate it? Is there some international OSM team, which goes to all the locations and checks, whether there is really a street? With Wikipedia articles it is simpler, because you can check sources mostly online or you have experts on subjects world-wide, not bound by location mostly. A map of the world seems like sooo much work to keep updated, while many Wikipedia articles can stay as they are, if they are OK now, because they treat historical topics for example, or mathematical things. There are new developments in those areas as well, but no one will take a wrong road when driving, because the Wikipedia article was not updated. Or at least it is not likely to happen. By introducing additional barriers, the update frequency on OSM might be even lower. I wonder how many contributors there are for OSM maps. I have personally never updated a map there and have been too lazy to read up on how to do that. One would have to find a clever way to validate map information, which does not inhibit participation.
- tpetry 6y ago
  The answer is crowd sourcing. Everything one person creates or modifies needs to be voted on by multiple people. After x votes and y percent in favour of the change it‘s visible to everyone. The x and y numbers could change depending on the user‘s reputation etc. Think of it like the stack overflow „tasks“ you are asked to do all the time: review this question, do this, do that ...
  - rathel 6y ago
    Because of the OSM data model, changes can't be really held in "unaccepted" state. One would run into editing conflicts[1]. On Wikipedia withholding edits until review works in some language versions [2], but because Wikipedia pages are "atoms" that stand on their own, they have it easier. [1] https://learnosm.org/en/josm/josm-conflict-resolution/ https://learnosm.org/en/josm/josm-conflict-resolution/ [2] https://en.wikipedia.org/wiki/Wikipedia:Flagged_revisions/Sighted_versions https://en.wikipedia.org/wiki/Wikipedia:Flagged_revisions/Si...
  - matkoniecz 6y ago
Semaphor 6y ago
It depends. In the outskirts of Lübeck, Germany, Google tried to tell me to walk through a private company area with a gate. OSM not only knew about this, it had far (far, far) more details about everything. In Waterkloof Heights, Pretoria, South Africa OSM was roughly 7 years out of date and had barely any information even then.
- SiempreViernes 6y ago
  When driving to a beach in Europe, google once tried routing me through a small two-story house. And possibly into a cliff wall, the house sort of blocked the view of how "the road" went from there...
- Mediterraneo10 6y ago
  > In Waterkloof Heights, Pretoria, South Africa OSM was roughly 7 years out of date and had barely any information even then. I did some mapping on the Garden Route years ago. When I later went and looked back at those communities, I was disappointed to see that little work had been done since my own edits long ago. And in general, much seemed to have been added by other foreign overlanders like myself, and not any kind of local OSM community. I wondered if this is because those demographics in South Africa that are affluent enough to be interested in nerdy tech things like OSM, have also grown up in a culture that always warned them of idly walking around towns with phone in hand?
  - Semaphor 6y ago
    Very possible. Waterkloof Heights is a very rich area, the only white people I’ve seen walking besides me were joggers. The black people were all there to work at the mansions.
  - jmkb 6y ago
    Meanwhile, tiny Lesotho (a landlocked enclave entirely within, and distinctly less affluent than, South Africa) has a passionately active OSM community uniting locals on the ground and international mappers who use satellite imagery. Last year it was officially declared[1] the first country to be "completely mapped" in OSM -- a moving target obviously, but impressive nonetheless. [1] https://maplesotho.wordpress.com/ https://maplesotho.wordpress.com/
Heliosmaster 6y ago
Anecdotally, I've always found really up-to-date information. Worst thing was that some rural areas had not a lot of data (so I added it, at least a bit).
AlexTrask 6y ago
I edited OSM for 10 years and in my opinion it depends of the zone. In my zone the comunity it's organized, the people checks strange edits and also have comunity projects. For example we detected that there was a lot of streets without name some years ago and we made tools to control the streets without name and worked on it, now about the 90% of the streets have a name. As I said for me the problem it's that the quality of information it's not homogenous
growlist 6y ago
You've hit on the central conundrum with OSM: it's a good enough proposition to make it interesting (free as in beer for many uses, acceptably accurate in many places etc), but a long long way from being a reliable, authoritative single source of truth. Which - let's be honest - is often true of Wikipedia as well, moderators notwithstanding. Improving OSM is an area of active research I believe, but I'd suggest the scale of the challenge is of a different magnitude than can be met by the type of ideas you mention. Think about something like Google Maps, or in the UK OS MasterMap. Each of these likely has hundreds of millions of R&D invested in tools, techniques, processes and infrastructure behind them, as well as an ongoing budget in the hundreds of millions, and numerous, well organised, skilled, full time staff. There's a critical mass behind such efforts that as yet OpenStreetMap has not been able to muster, which is no criticism of OSM, but just an acknowledgement of the reality of the situation. We're talking about setups that take dedicated organisations years to develop - the history of the Ordnance Survey for example goes back to 1747!
- matkoniecz 6y ago
  > authoritative single source of truth This is NOT a role for the OSM[0]. In many cases it is the best source of map data, but it will never be the sole single authoritative source of truth. Treating it as one is horrible idea and will end in tears. [0]or Google Maps
  - growlist 6y ago
    I agree, see my other comment - it's some organisations that have an expectation of OSM that is unreasonable.
  - kqr 6y ago
    Of course, a map will never be the authoritative source of a territory -- hence the saying. But what sets OSM apart from many other mapping projects is its focus on local knowledge and actual people. If people in an area call something by a name, then in OSM, that's its name, even if there's no official name for it, or if nobody really uses the official name.
matkoniecz 6y ago
This kind of statements are useless without specifying what kind of data was wrong and where. In my city (Kraków, Poland) road data, bicycle infrastructure, pedestrian infrastructure is extremely well mapped. Shops? Probably about 3% of them are mapped. China? You can find entire cities not mapped at all (for multiple reasons, starting from fact that mapping in China is illegal without permission from government). If you spot any wrong data (or very, very important missing data) - please report it https://wiki.openstreetmap.org/wiki/Notes#Adding_notes https://wiki.openstreetmap.org/wiki/Notes#Adding_notes Or even better, map it if you want - feel free to ask for help, for example here (or elsewhere, see https://wiki.openstreetmap.org/wiki/Contact_channels https://wiki.openstreetmap.org/wiki/Contact_channels https://community.osm.be/ https://community.osm.be/ )
gnufx 6y ago
If you want discussion, look at regional mail lists, perhaps, if the UK one is anything to go by. I don't share that impression of the data in the UK.

rathel 6y ago

Examples, please. There is bad info here and there, but it's minority (if you want 100% accuracy it's delusional, even Google Maps has a ton of crap, starting from skylights that AI recognized as buildings [3] to businesses you can't even tell they exist). Also, for a long time now each changeset can be discussed. [1] I have a hunch you may be from a country where the community has not took up too much [2]. These tend to have more dubious data. [1] https://blog.openstreetmap.org/2014/11/02/introducing-changeset-discussions/ https://blog.openstreetmap.org/2014/11/02/introducing-change... [2] https://osmstats.neis-one.org/?item=countries https://osmstats.neis-one.org/?item=countries [3] https://goo.gl/maps/knpT8eRR4YhB6Sm86 https://goo.gl/maps/knpT8eRR4YhB6Sm86 - look at the shopping mall in satellite and map view

ucosty 6y ago
Not OP, but I'm in the UK and find the OSM data to be not all that useful. The streets themselves seem mostly accurate, but there is almost no building data. If you had to rely on OSM data to get around, your best bet would be to live and work in northeast London, within the circular.
- Doctor_Fegg 6y ago
  YMMV really. OSM in the UK has orders of magnitude more cycling and walking data than Google, for example, but fewer buildings. I don't generally find I need buildings in my day-to-day mapping needs but everyone's different I guess.
  - bpye 6y ago
    I have found this in the U.K., Canada, Japan and Taiwan - OpenStreetMaps is way better as a pedestrian or a cyclist in my experience, at least in relatively built up areas.
- the_pwner224 6y ago
  Its the same in many parts of the USA. For example in a college town with 50k students / 50k permanent population, streets are complete with sidewalks and bike trails etc., but good luck trying to find nearby restaurants by looking around the map, like you can do with google maps. Even the buildings which are marked with their name and type usually lack operating hours, phone number, website url and other sorts of info that GMaps has. But the situation is much better - close to GMaps level - in big cities here.
  - matkoniecz 6y ago
    > lack operating hours If you have an Android phone - I want to plug StreetComplete app here. I never mapped opening hours before I started using it. It asks simple questions, there is no requirement to learn anything OSM specific. Registration at OSM website is probably the biggest hurdle. (disclaimer: I am one of contributors to that project)
- smichel17 6y ago
  A great thing about OSM is that you can improve the data, specifically in your local area so that at the very least it will no longer have spotty coverage of the places you frequent. I use https://github.com/westnordost/StreetComplete https://github.com/westnordost/StreetComplete for doing this so I don't have to make the investment to learn the technical details of tags.
  - drstewart 6y ago
    "You have to create the map to use the map" isn't as compelling of a feature for a mapping service as you think it is
    - nelgaard 6y ago
      OP talked about improving OSM using StreetComplete. So this is about adding missing speed limits, opening hours, roof shapes, recycling containers, tactile pavement at bus stops, etc. Hardly "creating the map". When I travel to a place where OSM could need improvement I make sure that at least the hotel I will be staying at, nearby restaurants, bars and museums, etc are correctly tagged with phone numbers, opening hours etc. It is not much work and it gives me an opportunity to decide what I want to do before I leave. When I get there I then have everything ready offline in OsmAnd. In a way I just update OSM instead of adding my own bookmarks/favorites.
    - smichel17 6y ago
      nelgaard got the gist of it. I've never encountered an area where OSM didn't have the roads and buildings already mapped. I have encountered areas where it didn't have the speed limits and house numbers of those structures (and, once or twice, places lacking a road name, but that's pretty rare). But, my personal perspective was a little different. It's more like: I want to use OSM, and get others to use OSM instead of proprietary mapping solutions like Google's, for moral reasons. In that context, being able to make sure my friends can always get to my house using OSM is a great boon. I realize I didn't share that perspective in my original post, which is probably the source of the confusion here.
- 7952 6y ago
  The great thing about OSM is the currency. New developments will be added more qucikly. But for completeness and consistency Ordnance Survey data will just be better. And they do have a free building dataset, and even pre-generated vector tiles.
  - gnufx 6y ago
    Even the OS data -- at least what I can see in Aurora -- might not be as complete as where someone has taken an interest in a new development, such as one near here. The OpenData building shapes are a good start, but they appear to have been auto-generated, and usually need correcting from up-to-date satellite imagery, if not a survey.
- gnufx 6y ago
  Whether buildings are mapped, and in what detail, just depends on the area; detailed mapping definitely isn't confined to London. The main utility of mapping buildings for navigation might be to tag them with postcodes, even if that's not ideal. Roads are typically at least as accurate as OS OpenData where I've looked.
c0nsumer 6y ago
Roads in rural areas are a HUGE problem. Particular in remote / forested places where logging roads were used in TIGER data. For example, places like this: https://www.openstreetmap.org/#map=14/46.6397/-87.7149 https://www.openstreetmap.org/#map=14/46.6397/-87.7149 There's a number of us who use this kind of data for building back country cycling routes, and while it's great guidance, there's both the problem with roads being listed that are no longer passable (nor often even visible at ground level), or other routes available that don't exist. Of course, this gets back to the way much of the data is crowd sourced... And we're working on that. :)
- gerbal 6y ago
  Anecdotally, in Northern Wisconsin, GMaps is only slightly better than OSM on abandoned logging roads. There are a lot of private gated or abandoned roads and logging tracks which Google Maps will use for route planning.
  - c0nsumer 6y ago
    That's kinda how it is here. The best thing seems to be to bounce between Google Maps and OSM and an aerial view to get a sense of what's up. For an interesting example, check out this place: Google: https://www.google.com/maps/@46.4133003,-87.5248774,5345m/data=!3m1!1e3 https://www.google.com/maps/@46.4133003,-87.5248774,5345m/da... OSM: https://www.openstreetmap.org/#map=14/46.4161/-87.5406 https://www.openstreetmap.org/#map=14/46.4161/-87.5406 Neither has the actual current lake boundaries (as used as a retention pond), Google has roads where the lake is while OSM doesn't, etc.
    - maxerickson 6y ago
      3dep can be a huge help in figuring out how prominent tree covered roads are. https://elevation.nationalmap.gov/arcgis/rest/services/3DEPElevation/ImageServer https://elevation.nationalmap.gov/arcgis/rest/services/3DEPE... It's listed as a Theme Overlay at https://viewer.nationalmap.gov/services/ https://viewer.nationalmap.gov/services/
INTPenis 6y ago
I wanted to use node data on shops, like if they take cash or not for example. Noticed there was already a lot of shops with data in there. But none of it was accurate. Just one example is that an Apple store in Sweden was set to accept cash. Just for fun I called them up and asked if they take cash and they laughed at me. It seemed to me that someone had automated input for businesses without actually verifying their data.

zelphirkalt 6y ago

It sounds like a good plan, however, there is the question of how you want to validate the map information. Someone adds a street. What do you want to do to validate it? Is there some international OSM team, which goes to all the locations and checks, whether there is really a street? With Wikipedia articles it is simpler, because you can check sources mostly online or you have experts on subjects world-wide, not bound by location mostly. A map of the world seems like sooo much work to keep updated, while many Wikipedia articles can stay as they are, if they are OK now, because they treat historical topics for example, or mathematical things. There are new developments in those areas as well, but no one will take a wrong road when driving, because the Wikipedia article was not updated. Or at least it is not likely to happen. By introducing additional barriers, the update frequency on OSM might be even lower. I wonder how many contributors there are for OSM maps. I have personally never updated a map there and have been too lazy to read up on how to do that. One would have to find a clever way to validate map information, which does not inhibit participation.

tpetry 6y ago
The answer is crowd sourcing. Everything one person creates or modifies needs to be voted on by multiple people. After x votes and y percent in favour of the change it‘s visible to everyone. The x and y numbers could change depending on the user‘s reputation etc. Think of it like the stack overflow „tasks“ you are asked to do all the time: review this question, do this, do that ...
- rathel 6y ago
  Because of the OSM data model, changes can't be really held in "unaccepted" state. One would run into editing conflicts[1]. On Wikipedia withholding edits until review works in some language versions [2], but because Wikipedia pages are "atoms" that stand on their own, they have it easier. [1] https://learnosm.org/en/josm/josm-conflict-resolution/ https://learnosm.org/en/josm/josm-conflict-resolution/ [2] https://en.wikipedia.org/wiki/Wikipedia:Flagged_revisions/Sighted_versions https://en.wikipedia.org/wiki/Wikipedia:Flagged_revisions/Si...
- matkoniecz 6y ago

Semaphor 6y ago

It depends. In the outskirts of Lübeck, Germany, Google tried to tell me to walk through a private company area with a gate. OSM not only knew about this, it had far (far, far) more details about everything. In Waterkloof Heights, Pretoria, South Africa OSM was roughly 7 years out of date and had barely any information even then.

SiempreViernes 6y ago
When driving to a beach in Europe, google once tried routing me through a small two-story house. And possibly into a cliff wall, the house sort of blocked the view of how "the road" went from there...
Mediterraneo10 6y ago
> In Waterkloof Heights, Pretoria, South Africa OSM was roughly 7 years out of date and had barely any information even then. I did some mapping on the Garden Route years ago. When I later went and looked back at those communities, I was disappointed to see that little work had been done since my own edits long ago. And in general, much seemed to have been added by other foreign overlanders like myself, and not any kind of local OSM community. I wondered if this is because those demographics in South Africa that are affluent enough to be interested in nerdy tech things like OSM, have also grown up in a culture that always warned them of idly walking around towns with phone in hand?
- Semaphor 6y ago
  Very possible. Waterkloof Heights is a very rich area, the only white people I’ve seen walking besides me were joggers. The black people were all there to work at the mansions.
- jmkb 6y ago
  Meanwhile, tiny Lesotho (a landlocked enclave entirely within, and distinctly less affluent than, South Africa) has a passionately active OSM community uniting locals on the ground and international mappers who use satellite imagery. Last year it was officially declared[1] the first country to be "completely mapped" in OSM -- a moving target obviously, but impressive nonetheless. [1] https://maplesotho.wordpress.com/ https://maplesotho.wordpress.com/

Heliosmaster 6y ago

Anecdotally, I've always found really up-to-date information. Worst thing was that some rural areas had not a lot of data (so I added it, at least a bit).

AlexTrask 6y ago

I edited OSM for 10 years and in my opinion it depends of the zone. In my zone the comunity it's organized, the people checks strange edits and also have comunity projects. For example we detected that there was a lot of streets without name some years ago and we made tools to control the streets without name and worked on it, now about the 90% of the streets have a name. As I said for me the problem it's that the quality of information it's not homogenous

growlist 6y ago

You've hit on the central conundrum with OSM: it's a good enough proposition to make it interesting (free as in beer for many uses, acceptably accurate in many places etc), but a long long way from being a reliable, authoritative single source of truth. Which - let's be honest - is often true of Wikipedia as well, moderators notwithstanding. Improving OSM is an area of active research I believe, but I'd suggest the scale of the challenge is of a different magnitude than can be met by the type of ideas you mention. Think about something like Google Maps, or in the UK OS MasterMap. Each of these likely has hundreds of millions of R&D invested in tools, techniques, processes and infrastructure behind them, as well as an ongoing budget in the hundreds of millions, and numerous, well organised, skilled, full time staff. There's a critical mass behind such efforts that as yet OpenStreetMap has not been able to muster, which is no criticism of OSM, but just an acknowledgement of the reality of the situation. We're talking about setups that take dedicated organisations years to develop - the history of the Ordnance Survey for example goes back to 1747!

matkoniecz 6y ago
> authoritative single source of truth This is NOT a role for the OSM[0]. In many cases it is the best source of map data, but it will never be the sole single authoritative source of truth. Treating it as one is horrible idea and will end in tears. [0]or Google Maps
- growlist 6y ago
  I agree, see my other comment - it's some organisations that have an expectation of OSM that is unreasonable.
- kqr 6y ago
  Of course, a map will never be the authoritative source of a territory -- hence the saying. But what sets OSM apart from many other mapping projects is its focus on local knowledge and actual people. If people in an area call something by a name, then in OSM, that's its name, even if there's no official name for it, or if nobody really uses the official name.

matkoniecz 6y ago

This kind of statements are useless without specifying what kind of data was wrong and where. In my city (Kraków, Poland) road data, bicycle infrastructure, pedestrian infrastructure is extremely well mapped. Shops? Probably about 3% of them are mapped. China? You can find entire cities not mapped at all (for multiple reasons, starting from fact that mapping in China is illegal without permission from government). If you spot any wrong data (or very, very important missing data) - please report it https://wiki.openstreetmap.org/wiki/Notes#Adding_notes https://wiki.openstreetmap.org/wiki/Notes#Adding_notes Or even better, map it if you want - feel free to ask for help, for example here (or elsewhere, see https://wiki.openstreetmap.org/wiki/Contact_channels https://wiki.openstreetmap.org/wiki/Contact_channels https://community.osm.be/ https://community.osm.be/ )

gnufx 6y ago

If you want discussion, look at regional mail lists, perhaps, if the UK one is anything to go by. I don't share that impression of the data in the UK.