7 ms·

They could sponsor OpenStreetMaps and dedicate resources to it, donate satellite imagery etc. They'd get a lot of great data in return.

by untog 8y ago

macintux 8y ago
OSM is orders of magnitude worse than Apple's current data set in my area, and I doubt that's unusual. So the question for Apple is why tie themselves to a 3rd party platform? If they're going to go to this massive level of effort to generate the best data set they can achieve, they'll want to control the full stack for quality and predictability. They could feed the OSM-relevant data they create to the platform, but that means a lot more work to create and maintain the integration tools and manage the workflow, along with potentially reduced revenues if they make OSM too good. I doubt they'll care about any threat from OSM, but I doubt they'd see the overall effort as a good business investment. This isn't as straightforward as open sourcing some software.
- saturn_vk 8y ago
  You could fix your spotty coverage yourself. The areas I've been at had exceptional detail.
  - macintux 8y ago
    I spent some time starting to do so, but the gaps are cavernous, and I decided I had I other priorities.
  - __sr__ 8y ago
    Yeah, I’ve nothing better to do than roam around fixing maps. It’s not like I have a daily job or anything. /s It the same argument when someone talks about an issue with an open source software. The response is, “The source code is available, go fix it yourself.” It doesn’t work that way for anything but trivial issues.
    - 394549 8y ago
      > Yeah, I’ve nothing better to do than roam around fixing maps. It’s not like I have a daily job or anything. /s No one said it has to be your job or passion. I've fixed a couple of areas in OSM, but I've probably spent no more than a few hours working on it in total. > It the same argument when someone talks about an issue with an open source software. The response is, “The source code is available, go fix it yourself.” It doesn’t work that way for anything but trivial issues. Fixing a map for a locale you're familiar with is orders of magnitude easier than fixing a bug in an unfamiliar codebase potentially built with unfamiliar technology. That's assuming you know how to code in the first place.
      - __sr__ 8y ago
        > No one said it has to be your job or passion. I've fixed a couple of areas in OSM, but I've probably spent not more than a few hours working on it in total. That doesn’t work unless everyone else is doing it. And for that to work, everyone should be using it in the first place. It’s a catch 22 problem. To be fair, it is possible that people do this where you live - which may be a reason why it is usable there, but I have not found that to be the case at the places I have visited. Perhaps it is not as complicated as fixing a software bug, but the idea is same - the data is available publically, go fix it.
        macintux 8y ago
        It's not as complicated as fixing a software bug, but it's not trivial for the first-time user either. Create an account, figure out how to navigate the interface, figure out what rules apply to your update, apply update, get reprimanded because you followed the explicit rule for "I'm not quite sure what to do with this".
        394549 8y ago
        > That doesn’t work unless everyone else is doing it. And for that to work, everyone should be using it in the first place. It’s a catch 22 problem. If that were actually true, Wikipedia would be no more than four stub articles written by Jimmy Wales. The real barrier is people looking at a volunteer project and asking themselves "how can I take from this what I want?" rather than "how can I contribute to this, even in a small way?" No one says anyone has to take on the herculean task of totally fixing all the problems they see in their area. But if you spend a little time fixing something, it will get better and perhaps even attain a high enough quality to satisfy you.
- pythonaut_16 8y ago
  I think it's kinda like Google working with Android as an open source project. Google makes money on Android primarily by using Android as a means of delivering the Google ecosystem. They don't care that it's open source because they don't make money by selling it. Apple could do the same thing with maps. Support a robust open mapping initiative to make sure their primary business (selling iPhones and other devices) has a solid platform for emerging technologies like AR. Making sure they have a solid maps offering supports the Apple (aka iPhone) ecosystem. Also similarly to Google, they can still push plenty of proprietary things on top of the open parts of the initiative.
  - borgel 8y ago
    Plus it seems like this would weaken Google's partial stranglehold on the market is a side effect.
    - oblio 8y ago
      Commoditize your complement.
      - mikejarema 8y ago
        I like that line, is it yours or does it have some further history/examples behind it?
        mkl 8y ago
        It's been around for a while. Here it is talked about by Joel Spolsky in 2002: https://www.joelonsoftware.com/2002/06/12/strategy-letter-v/ https://www.joelonsoftware.com/2002/06/12/strategy-letter-v/ Gwern's article about it was recently discussed here on HN: https://news.ycombinator.com/item?id=17047348 https://news.ycombinator.com/item?id=17047348
  - gxs 8y ago
    This is a great idea. I hope someone at Apple is reading. Makes perfect sense, they don't need to make money on it, just an alternative to google.
- Doctor_Fegg 8y ago
  > OSM is orders of magnitude worse than Apple's current data set in my area, and I doubt that's unusual. As ever, define "worse" and "my area". OSM is consistently worse worldwide (compared to Google, TomTom and Here) for geocoding. OSM is consistently worse for lane guidance. OSM is generally worse than Google worldwide for commercial POIs (shops, restaurants etc.), but on a par with TomTom and Here, and this is very location-dependent. OSM is better worldwide, by several orders of magnitude, for pedestrian and bicycle mapping. OSM is generally better for non-commercial POIs. OSM is very often better in all aspects (save geocoding) in rural locations, particularly in Europe. If you're an American motorist then Google is your best map right now. If you're a European cyclist then OSM is really the only game in town. It's not a black-and-white issue.
  - yetanother1980 8y ago
    I was about to respond to the GP asking for details because where I am and as mainly a pedestrian and cyclist OSM is really the only decent solution (unless you like walking 30minutes in a large circle rather than 5).
- rubyn00bie 8y ago
  I think Apple sponsoring and improving OSM would actually be the smartest and best way out of this debacle. I think a perfect example of Apple doing this is: Webkit. Webkit gets back a lot from Apple (though I'm sure there's plenty of room to be miffed at Apple for being slow sometimes, but it still doesn't disprove the point) and it's a core part of their user-facing applications. Personally, I'd like to see them just open source what they have and let the world enjoy. Either approach would certainly remove the death grip Google has on mapping solutions right now. Edit: update for grammar and clarity fixes.
  - krick 8y ago
    > Apple sponsoring and improving OSM would actually be the smartest and best way out of this debacle I like to be thinking that, but being aware of me liking that thought, I'm asking myself: would I actually do so being some Apple executive? And I'm not sure I would. I do think that being community-supported is really, really great, and OSM has a pretty good community too. I'm a big opensource proponent, so it's easy for me to say, but I also think having their maps (software and data) free (as in GNU, yeah) wouldn't harm Apple, given the map is not their unique product they are selling. And being a native supporter of something that could be wikipedia for maps would be huge. So, yeah… yay OSM? But, with all that said: IMO, OSM is a huge organizational mess. Giving them money & satellites, and then just expecting it will all "solve itself out" isn't really something I would do being responsible for Apple Maps. For starters, it's not like all OSM community members automatically become Apple employees, you're basically donating to somebody over whom you won't have any leverage. What's more important, it's not clear what it the goal of the donation, what OSM is expected to do with that? And how it will serve Apple? Their priorities seem to be quite off, OSM doesn't give a shit about any "product" whatsoever, they are just guys passionate about collecting all sorts of geo-data into databases nobody ever uses (at least, as far as OSM is concerned, they are the database: they don't make apps). And it's even questinable if such way of collecting data is the way to go in the age of imaging satellites, widely accepted spyware on every phone, ML and "big data". And it's not like this is cheap, any way it would be done. And, no offense, but "partnering" with OSM wouldn't be exactly partnering. OSM doesn't have managers, engineers or cartographers any better than Apple can hire, all they have is community, which wouldn't even be OSM community anymore, if Apple rolls out a better (and more accessible) open geo-spatial DB than OSM has. Which, I believe, they are capable of.
- hrktb 8y ago
  For me the issue is more that they don’t seem to be able to pull it off. Sure if they could have a super high quality dataset they build themselves it would be stupid to share it, but they don’t have that data, and I wouldn’t bet on them to have it even putting all their weight in the effort. There is also the issue of motivation for the user to give feedback. If I do it in Google Maps I know it will benefit most users. In OSM it’s a pure gift to the community. Apple Maps ? it’s less clearcut.
- untog 8y ago
  The key part in what you said is "my area". There are a great many areas in which Apple Maps is fantastic - usually well-populated cities or surrounding areas in the US. The reason to partner with OSM would be to cover the other 99% of landmass. It's very, very difficult to scale an operation like that otherwise.
simonh 8y ago
That would massively benefit OSM, and if OSM became Apple Maps/Google Maps calibre resource it would also benefit other OSM users including potentially Apple's competitors, but how would it benefit Apple? Bear in mind if malicious entries got into OSM and made their way into Apple's data, guess who eats the smelly sandwitch for that?
reaperducer 8y ago
They could sponsor OpenStreetMaps and dedicate resources to it According to the article, OpenStreetMaps was one of the data sources that Apple used and decided wasn't up to snuff. The article doesn't specify what the problems were with OSM, or if there were any problems unique to OSM, but one of Apple's priorities is rapid map updates, which is something that can be accomplished faster if it's in-house. In addition, Apple intends to update its new maps in real-time (new road opens and 100 iPhone users drive on it, it magically appears on the map). I don't think that's possible with OSM.
- alex_duf 8y ago
  That's absolutely possible on osm, in fact new roads are often on OSM before they reach Google maps
  - StudentStuff 8y ago
    Google Maps gets really bad in rural areas, I was shocked on a recent trip in Washington state at how bad their maps are. They seemed to have half the businesses in town on the map, but the roads were inaccurate, missing, and sometimes just plain wrong. OSM comparatively was of much higher caliber, which was surprising.
  - isostatic 8y ago
    My estate is 3 years old. Google has no mention. Apple is incorrect. Openstreetmap is accurate since the day each road was opened up.
- Doctor_Fegg 8y ago
  Apple has a team right now working on OSM: https://github.com/osmlab/appledata/issues https://github.com/osmlab/appledata/issues The article just strikes me as plain wrong in this regard. Yes, of course Apple isn't going to use OSM in the Bay Area, but it's a big world out there.
- maxerickson 8y ago
  There's no technical limitation whatsoever in the way of using OSM as a primary data source and also doing things like using phone telemetry to add roads to your local display map. It'd likely be necessary to make that data available under the same license OSM is under, but that'd be it. The way people use OSM, they grab a dump of the data and then load it into a database or whatever (probably processing it into a convenient schema) and then build maps from that database. So it's pretty easy to have more than 1 source contributing to the map you are displaying. For updates, edits to OSM are available minute by minute, so if your data ingestion is up to it you don't have to put up with the stale data: https://wiki.openstreetmap.org/wiki/Planet.osm/diffs#Minutely.2C_Hourly.2C_or_Daily_diffs https://wiki.openstreetmap.org/wiki/Planet.osm/diffs#Minutel...
briandear 8y ago
Why would Apple want to base their business on their he competency of OpenStreetMaps. As a shareholder, I would find that alarming; as a user, Apple Maps is far better as it is right now.
rdlecler1 8y ago
Agree, Apple should just commoditized maps with a big open source initiative to take the leverage/power away from Google. They just need a highly functional solution for their platform.
heed 8y ago
Apple already contributes to and works with OSM data.
madeofpalk 8y ago
Fun fact: they do! https://github.com/osmlab/appledata https://github.com/osmlab/appledata e.g. https://lists.openstreetmap.org/pipermail/talk-au/2018-May/011846.html https://lists.openstreetmap.org/pipermail/talk-au/2018-May/0... > * I work for Apple’s Maps team. We are interested in doing some fixes and improvements to inland water features in Australia on OSM, such as adding and improving geometry of polygons for lakes and wide rivers, fixing broken relations, and correcting alignment issues for inland features when they meet the coast.* > Atlas [1], a tool Apple created for querying, visualizing, and storing OpenStreetMap data. > Data Improvement Projects [2] being worked on by the Apple Data Team. > Building Footprint Data [3] that Apple is sharing. [1]: https://github.com/osmlab/atlas https://github.com/osmlab/atlas [2]: https://github.com/osmlab/appledata/issues https://github.com/osmlab/appledata/issues [3]: https://github.com/osmlab/appledata/blob/master/BUILDINGS.md https://github.com/osmlab/appledata/blob/master/BUILDINGS.md
toasterlovin 8y ago
Yeah, but they’re actually collecting all this new data for autonomous vehicles. Improvements to their maps app is just a freebee. For autonomous cars, mapping data this detailed is probably a several billion dollar competitive advantage that they don’t want to give away for free.

They could sponsor OpenStreetMaps and dedicate resources to it, donate satellite imagery etc. They'd get a lot of great data in return.

by untog 8y ago

They could sponsor OpenStreetMaps and dedicate resources to it, donate satellite imagery etc. They'd get a lot of great data in return.

macintux 8y ago

OSM is orders of magnitude worse than Apple's current data set in my area, and I doubt that's unusual. So the question for Apple is why tie themselves to a 3rd party platform? If they're going to go to this massive level of effort to generate the best data set they can achieve, they'll want to control the full stack for quality and predictability. They could feed the OSM-relevant data they create to the platform, but that means a lot more work to create and maintain the integration tools and manage the workflow, along with potentially reduced revenues if they make OSM too good. I doubt they'll care about any threat from OSM, but I doubt they'd see the overall effort as a good business investment. This isn't as straightforward as open sourcing some software.

saturn_vk 8y ago
You could fix your spotty coverage yourself. The areas I've been at had exceptional detail.
- macintux 8y ago
  I spent some time starting to do so, but the gaps are cavernous, and I decided I had I other priorities.
- __sr__ 8y ago
  Yeah, I’ve nothing better to do than roam around fixing maps. It’s not like I have a daily job or anything. /s It the same argument when someone talks about an issue with an open source software. The response is, “The source code is available, go fix it yourself.” It doesn’t work that way for anything but trivial issues.
  - 394549 8y ago
    > Yeah, I’ve nothing better to do than roam around fixing maps. It’s not like I have a daily job or anything. /s No one said it has to be your job or passion. I've fixed a couple of areas in OSM, but I've probably spent no more than a few hours working on it in total. > It the same argument when someone talks about an issue with an open source software. The response is, “The source code is available, go fix it yourself.” It doesn’t work that way for anything but trivial issues. Fixing a map for a locale you're familiar with is orders of magnitude easier than fixing a bug in an unfamiliar codebase potentially built with unfamiliar technology. That's assuming you know how to code in the first place.
    - __sr__ 8y ago
      > No one said it has to be your job or passion. I've fixed a couple of areas in OSM, but I've probably spent not more than a few hours working on it in total. That doesn’t work unless everyone else is doing it. And for that to work, everyone should be using it in the first place. It’s a catch 22 problem. To be fair, it is possible that people do this where you live - which may be a reason why it is usable there, but I have not found that to be the case at the places I have visited. Perhaps it is not as complicated as fixing a software bug, but the idea is same - the data is available publically, go fix it.
      - macintux 8y ago
        It's not as complicated as fixing a software bug, but it's not trivial for the first-time user either. Create an account, figure out how to navigate the interface, figure out what rules apply to your update, apply update, get reprimanded because you followed the explicit rule for "I'm not quite sure what to do with this".
      - 394549 8y ago
        > That doesn’t work unless everyone else is doing it. And for that to work, everyone should be using it in the first place. It’s a catch 22 problem. If that were actually true, Wikipedia would be no more than four stub articles written by Jimmy Wales. The real barrier is people looking at a volunteer project and asking themselves "how can I take from this what I want?" rather than "how can I contribute to this, even in a small way?" No one says anyone has to take on the herculean task of totally fixing all the problems they see in their area. But if you spend a little time fixing something, it will get better and perhaps even attain a high enough quality to satisfy you.
pythonaut_16 8y ago
I think it's kinda like Google working with Android as an open source project. Google makes money on Android primarily by using Android as a means of delivering the Google ecosystem. They don't care that it's open source because they don't make money by selling it. Apple could do the same thing with maps. Support a robust open mapping initiative to make sure their primary business (selling iPhones and other devices) has a solid platform for emerging technologies like AR. Making sure they have a solid maps offering supports the Apple (aka iPhone) ecosystem. Also similarly to Google, they can still push plenty of proprietary things on top of the open parts of the initiative.
- borgel 8y ago
  Plus it seems like this would weaken Google's partial stranglehold on the market is a side effect.
  - oblio 8y ago
    Commoditize your complement.
    - mikejarema 8y ago
      I like that line, is it yours or does it have some further history/examples behind it?
      - mkl 8y ago
        It's been around for a while. Here it is talked about by Joel Spolsky in 2002: https://www.joelonsoftware.com/2002/06/12/strategy-letter-v/ https://www.joelonsoftware.com/2002/06/12/strategy-letter-v/ Gwern's article about it was recently discussed here on HN: https://news.ycombinator.com/item?id=17047348 https://news.ycombinator.com/item?id=17047348
- gxs 8y ago
  This is a great idea. I hope someone at Apple is reading. Makes perfect sense, they don't need to make money on it, just an alternative to google.
Doctor_Fegg 8y ago
> OSM is orders of magnitude worse than Apple's current data set in my area, and I doubt that's unusual. As ever, define "worse" and "my area". OSM is consistently worse worldwide (compared to Google, TomTom and Here) for geocoding. OSM is consistently worse for lane guidance. OSM is generally worse than Google worldwide for commercial POIs (shops, restaurants etc.), but on a par with TomTom and Here, and this is very location-dependent. OSM is better worldwide, by several orders of magnitude, for pedestrian and bicycle mapping. OSM is generally better for non-commercial POIs. OSM is very often better in all aspects (save geocoding) in rural locations, particularly in Europe. If you're an American motorist then Google is your best map right now. If you're a European cyclist then OSM is really the only game in town. It's not a black-and-white issue.
- yetanother1980 8y ago
  I was about to respond to the GP asking for details because where I am and as mainly a pedestrian and cyclist OSM is really the only decent solution (unless you like walking 30minutes in a large circle rather than 5).
rubyn00bie 8y ago
I think Apple sponsoring and improving OSM would actually be the smartest and best way out of this debacle. I think a perfect example of Apple doing this is: Webkit. Webkit gets back a lot from Apple (though I'm sure there's plenty of room to be miffed at Apple for being slow sometimes, but it still doesn't disprove the point) and it's a core part of their user-facing applications. Personally, I'd like to see them just open source what they have and let the world enjoy. Either approach would certainly remove the death grip Google has on mapping solutions right now. Edit: update for grammar and clarity fixes.
- krick 8y ago
  > Apple sponsoring and improving OSM would actually be the smartest and best way out of this debacle I like to be thinking that, but being aware of me liking that thought, I'm asking myself: would I actually do so being some Apple executive? And I'm not sure I would. I do think that being community-supported is really, really great, and OSM has a pretty good community too. I'm a big opensource proponent, so it's easy for me to say, but I also think having their maps (software and data) free (as in GNU, yeah) wouldn't harm Apple, given the map is not their unique product they are selling. And being a native supporter of something that could be wikipedia for maps would be huge. So, yeah… yay OSM? But, with all that said: IMO, OSM is a huge organizational mess. Giving them money & satellites, and then just expecting it will all "solve itself out" isn't really something I would do being responsible for Apple Maps. For starters, it's not like all OSM community members automatically become Apple employees, you're basically donating to somebody over whom you won't have any leverage. What's more important, it's not clear what it the goal of the donation, what OSM is expected to do with that? And how it will serve Apple? Their priorities seem to be quite off, OSM doesn't give a shit about any "product" whatsoever, they are just guys passionate about collecting all sorts of geo-data into databases nobody ever uses (at least, as far as OSM is concerned, they are the database: they don't make apps). And it's even questinable if such way of collecting data is the way to go in the age of imaging satellites, widely accepted spyware on every phone, ML and "big data". And it's not like this is cheap, any way it would be done. And, no offense, but "partnering" with OSM wouldn't be exactly partnering. OSM doesn't have managers, engineers or cartographers any better than Apple can hire, all they have is community, which wouldn't even be OSM community anymore, if Apple rolls out a better (and more accessible) open geo-spatial DB than OSM has. Which, I believe, they are capable of.
hrktb 8y ago
For me the issue is more that they don’t seem to be able to pull it off. Sure if they could have a super high quality dataset they build themselves it would be stupid to share it, but they don’t have that data, and I wouldn’t bet on them to have it even putting all their weight in the effort. There is also the issue of motivation for the user to give feedback. If I do it in Google Maps I know it will benefit most users. In OSM it’s a pure gift to the community. Apple Maps ? it’s less clearcut.
untog 8y ago
The key part in what you said is "my area". There are a great many areas in which Apple Maps is fantastic - usually well-populated cities or surrounding areas in the US. The reason to partner with OSM would be to cover the other 99% of landmass. It's very, very difficult to scale an operation like that otherwise.

simonh 8y ago

That would massively benefit OSM, and if OSM became Apple Maps/Google Maps calibre resource it would also benefit other OSM users including potentially Apple's competitors, but how would it benefit Apple? Bear in mind if malicious entries got into OSM and made their way into Apple's data, guess who eats the smelly sandwitch for that?

reaperducer 8y ago

They could sponsor OpenStreetMaps and dedicate resources to it According to the article, OpenStreetMaps was one of the data sources that Apple used and decided wasn't up to snuff. The article doesn't specify what the problems were with OSM, or if there were any problems unique to OSM, but one of Apple's priorities is rapid map updates, which is something that can be accomplished faster if it's in-house. In addition, Apple intends to update its new maps in real-time (new road opens and 100 iPhone users drive on it, it magically appears on the map). I don't think that's possible with OSM.

alex_duf 8y ago
That's absolutely possible on osm, in fact new roads are often on OSM before they reach Google maps
- StudentStuff 8y ago
  Google Maps gets really bad in rural areas, I was shocked on a recent trip in Washington state at how bad their maps are. They seemed to have half the businesses in town on the map, but the roads were inaccurate, missing, and sometimes just plain wrong. OSM comparatively was of much higher caliber, which was surprising.
- isostatic 8y ago
  My estate is 3 years old. Google has no mention. Apple is incorrect. Openstreetmap is accurate since the day each road was opened up.
Doctor_Fegg 8y ago
Apple has a team right now working on OSM: https://github.com/osmlab/appledata/issues https://github.com/osmlab/appledata/issues The article just strikes me as plain wrong in this regard. Yes, of course Apple isn't going to use OSM in the Bay Area, but it's a big world out there.
maxerickson 8y ago
There's no technical limitation whatsoever in the way of using OSM as a primary data source and also doing things like using phone telemetry to add roads to your local display map. It'd likely be necessary to make that data available under the same license OSM is under, but that'd be it. The way people use OSM, they grab a dump of the data and then load it into a database or whatever (probably processing it into a convenient schema) and then build maps from that database. So it's pretty easy to have more than 1 source contributing to the map you are displaying. For updates, edits to OSM are available minute by minute, so if your data ingestion is up to it you don't have to put up with the stale data: https://wiki.openstreetmap.org/wiki/Planet.osm/diffs#Minutely.2C_Hourly.2C_or_Daily_diffs https://wiki.openstreetmap.org/wiki/Planet.osm/diffs#Minutel...

briandear 8y ago

Why would Apple want to base their business on their he competency of OpenStreetMaps. As a shareholder, I would find that alarming; as a user, Apple Maps is far better as it is right now.

rdlecler1 8y ago

Agree, Apple should just commoditized maps with a big open source initiative to take the leverage/power away from Google. They just need a highly functional solution for their platform.

heed 8y ago

Apple already contributes to and works with OSM data.

madeofpalk 8y ago

Fun fact: they do! https://github.com/osmlab/appledata https://github.com/osmlab/appledata e.g. https://lists.openstreetmap.org/pipermail/talk-au/2018-May/011846.html https://lists.openstreetmap.org/pipermail/talk-au/2018-May/0... > * I work for Apple’s Maps team. We are interested in doing some fixes and improvements to inland water features in Australia on OSM, such as adding and improving geometry of polygons for lakes and wide rivers, fixing broken relations, and correcting alignment issues for inland features when they meet the coast.* > Atlas [1], a tool Apple created for querying, visualizing, and storing OpenStreetMap data. > Data Improvement Projects [2] being worked on by the Apple Data Team. > Building Footprint Data [3] that Apple is sharing. [1]: https://github.com/osmlab/atlas https://github.com/osmlab/atlas [2]: https://github.com/osmlab/appledata/issues https://github.com/osmlab/appledata/issues [3]: https://github.com/osmlab/appledata/blob/master/BUILDINGS.md https://github.com/osmlab/appledata/blob/master/BUILDINGS.md

toasterlovin 8y ago

Yeah, but they’re actually collecting all this new data for autonomous vehicles. Improvements to their maps app is just a freebee. For autonomous cars, mapping data this detailed is probably a several billion dollar competitive advantage that they don’t want to give away for free.