@SimonPoole , am I detecting a hint of you saying “they’ll use OSM data to bridge to their own skeleton, while simultaneously backpedaling into removing as much of OSM as they can, as fast as they can”?
That does chime a note for me considering what I see (an opaque, shadowy rollout). I don’t know how true it is, but I nod my head at those sentiments. “Borrow long enough to clone being alive, discard that host as quickly as possible and necessary.”
Welcome to the jungle. (Uh, jungles have elephants…).
Well even it is just petty cash for 3 of the 4 founding members, I don’t quite see companies spending $7-10 million/year ($40 million/year for the whole OMF) just to get better machine learned land cover data or buildings.
Particularly given that we just established that they are really only interested in car-centric stuff.
Contributing to OSM till they are ready makes the stop-gap they are using better and at the same time improves their labeling data set (for the machine learning) so win-win.
How I understood it is that the data will be free for everyone including companies. The membership fee is only to secure a spot in working groups that will be defining a data model and deciding which data to use. If a big company is planning to lean heavily on this data, they definitely would want to make sure that the format will fit their use case.
I am afraid I fail to see how this is a win for OSM. When they release the first version, I am sure the quality will be bad. But with better imagery and ai models the situation will rapidly improve and other type of data layers will be added. At some point companies and app developers might find it better to use tidy and consistent Overture than messy and impossible to improve OSM. I think at that point we are in trouble.
We are in trouble if OSM becomes (or at least is seen as) “impossible to improve”. That’s why I believe it’s important to keep listening to our existing and potential contributors to ensure we have tools and practices that allow individuals and companies to contribute good quality data to OSM.
Some relationship with Overture so that we hear from those users would be beneficial. I hope we (OSMF) establish something in way of membership or similar.
I, too, find “impossible to improve OSM” to be far too severe a statement. OK, the social / consensus aspects of the project (and that’s only a tip of the big iceberg of what I mean by that) do make it “messy” at times. The IETF (and ISO and United Nations, for that matter) could be called similarly “messy.” But we are not “impossible to improve.” Agreeing with @RobJN , we must keep listening. Both to individuals who contribute, who I think are the backbone of the project, but also companies and large organizations who align with our data (and maybe our goals, maybe not). Our goals must include “the best geo data, period” and the only way that happens is through constant improvement, which is and must be built into the DNA of how we map. We lose that if we become “impossible.”
What is happening, what is going to happen between Overture and OSM is still unfolding and will continue to do so. Here and now people are anticipating whether there will be a “too late for OSM to do anything about it” scenario. I don’t think we have enough focus to do that accurately now, although I’m glad we continue to try to better “pull” focus and better understand the inevitable relationship between us that will ensue. However, it isn’t that we are going to “hear from” Overture, it will be that we engage them, and we must do so deliberately. Part of OSM being in that more-assertive posture includes us talking about things as we do here, so, let’s keep up the good work. Even as the specifics of what lies ahead remains hazy today.
Overture might be a win, it might be a “meh,” it might be a threat. If the latter, it must be identified and a plan formulated. My eyes are open, but I’m still rubbing them into focus about all this.
I suspect that companies’ view of OSM as impossible to improve was the one of the main reasons they started with Overture project in the first place.
Yes there can be many ways how to understand “impossible to improve”, but in this context it means “to take a decisive action on a reasonable time scale”.
Since there are no set guidelines about what is consensus, how to measure it or how to achieve it, the community is effectively paralyzed. Even a simple operation of changing one water tag takes weeks of nowhere leading discussions. Like people suggesting waiting another 10 years for it to solve itself. And after all that effort that was put into fixing the river tagging issue there are still some that claim it was something wrong. And this is just one of many examples.
But here’s one that I spotted on TomTom Roadrunner imagery a couple of months back: OpenStreetMap. If you can’t see the “GPS trace”, it shows traffic as going from the Banyula Drive / Arunja Way intersection, out to the Pacific Motorway exit off-ramp, straight through a sound-barrier fence!
The easier way to compare OSM against TomTom is Map Compare, which has options for Bing Maps (which uses TomTom) and a number of other sources. Bing also uses TomTom for its newer Streetside imagery. (To confirm, simply look down at the car hood.)
" “Ready access to geospatial information has fueled the innovation of many technologies and products, benefitting organizations and communities around the world,” said Deane Kensok, Esri ArcGIS Content CTO. “Esri is committed to expanding access to ready-to-use map data and helping the global geospatial community who are building the next generation of location-based apps and solutions for tomorrow’s challenges.”
Members of Overture will provide data and technological contributions, combining their resources to create complete, accurate, and extensible real-world map data that is available under an open data license. As a member, Esri will help curate data from its extensive collection of community contributions assembled through its well-established Community Maps Program, which contributors have chosen to share as open data.
“Overture’s mission to build the best possible open map data includes incorporating high-quality geospatial data from cities and local governments. As part of a worldwide database, data from those authoritative sources can support geospatial applications for billions of end users,” said Jan Erik Solem, chairperson for the Overture Maps Foundation and engineering director of Maps at Meta. "Esri’s long commitment to open data and supporting federal, state, and local government mapping efforts makes them a great addition.”
Esri will continue to both encourage and enable the sharing of open data by authoritative sources and to help integrate that data into the Overture ecosystem in various ways. This will benefit all users of Overture data, including government organizations and their constituents that will be able to access the Overture data through Esri’s ArcGIS software and other services.
“Esri has a long history of collecting and curating geospatial data from authoritative and community sources to enable a wide variety of use cases,” said Michael Kopenec, general manager for AWS Geospatial. “They have been a valued mapping partner of Amazon Location Service, and we are pleased to welcome Esri as a member of the Overture Maps Foundation, where they will bring their valuable experience to the project and accelerate delivery of high-quality mapping data to developers.”"
POSTED ON FEBRUARY 7, 2023
Improving Meta’s global maps
"… In 2023, we are looking forward to incorporating new open map data coming from the recently announced Overture Maps Foundation, which Meta co-founded along with Microsoft, AWS, and TomTom.
In some regions, we’ve introduced AI-based translations to increase the readability of map labels, like street names, for people around the world. We’ve also started introducing map features from open data sources. In select cities high resolution pavement data, tree locations, and even pavement markings bring an unparalleled level of detail to our maps. …"
… We’re also continuing to work with partners and communities to improve our maps by building reliable, open 3D geospatial map data for the metaverse and digital twins. …
"We parsed OSM’s complicated building and building:part tags to refashion our building features from the ground up. "
But they think that parsing that is complicated though? Maybe they also intend to render roof shapes.
What this has to do with overturemaps though I don’t know. The only thing that in my mind would somewhat simplify it for data consumers is if they’d directly prepare simple 3d models for each building. I think this is what OsmBuildings (and Cesium?) are doing?
