And just for your notes when you are dealing with them the next time: Development Seed that built their business off of OSM, are now an Overture member (and notably -not- of the OSMF).
Overture published a “beta” dataset in April:
Some interesting discussion on the Places dataset in their Github: Import OpenStreetMap places · Issue #96 · OvertureMaps/data · GitHub
Most users of Places data combine data from various sources to get to the coverage they want. We anticipate that users will combine Overture Places data with other sources, including commercially licensed data until a time when the open data is sufficient on its own. The CDLA license allows them to do that.
I guess there’s still no corporate editing of POIs in OSM
Just on that point - there is (and has been since basically forever) widespread corporate editing of POIs on OSM - in the vast majority of cases to the benefit of both OSM and the corporate editors.
Who’s doing that, and is it widespread? I couldn’t find any when I reviewed it, besides some ad-hoc updates from TomTom.
I posted a question in their Schema discussions whether they could add opening_hours as a field to their Place data. This would be quite useful for potential imports to OSM as many small businesses use Facebook as their primary website and keep opening hours updated there. Unfortunately no replies
Overture has now gone out of beta for all their themes except “Transport”, and they’re launching a new “adresses” theme based on OpenAdresses
It’s now also possible to browse the dataset here: https://explore.overturemaps.org/
For starters, there are some SEO companies, mainly in the USA, that spam info from their clients all over the internet, including a POI on OSM. Usually these POIs are nodes by newly made accounts that have the name of the client company. Those accounts are subsequently abandoned. The data is usually of mediocre quality, but it’s reliable data and it’s good enough that we can polish it up a bit and end up with decent POIs.
LWN has an article on Overture, presently paywalled but it will be freely available in two weeks time.
It seems to rely on people being too stupid to realise that it was paid for by the linux foundation aka Overture.
The article is now freely available.
Interesting comments, opening hours will never be part of the Places dataset as it’s too volatile. Aka you need to pay Meta for access to it?
The expactation is also that Overture contains “less” OSM over time…
The road network, while constantly changing, is relatively easy to keep up with; the same is true of addresses and such. This kind of data is manageable as a shared base layer that all can contribute to and use. The attached layers, though, which might include traffic data, restaurant reviews, opening hours, or any other sort of add-on data, are far too volatile to be managed in an open layer. This data will, he said, continue to be a proprietary product indefinitely.
The companies that launched Overture Maps first tried hard to turn OpenStreetMap into the sort of project they needed, he said; the OpenStreetMap contributors “successfully fended that off”. So Overture Maps was created with a focus on the end users of the data rather than on the contributors. While a bit over half of the project’s data comes from OpenStreetMap now, he said, he expects that proportion to fall in the future.
I think the way they put it previously was that the proportion of OSM data would fall relative to the total amount of data they offer. In other words, they expect the amount of OSM data to remain stable while the amount of non-OSM data will go up.
Mapillary has switched to a background map that seems to combine OSM streets/buildings/landuse and Overture POIs, or maybe it’s completely made from Overture data?
The map style is called “Facebook Canterbury 1.0”.
Not sure when they made the switch, but I just noticed it recently.
To be clear, they hate the ODbL. As soon as they get a road network under a different license, they will switch to it.
In the US, the TIGER data set is public domain and updates at least annually, and it includes all roads nationwide. I would assume there are other road network data sets that can be gotten for free in other countries also. TIGER certainly is or underlies what Google Maps is using as far as I can tell.
I imagine this would be a different story in developing nations where OSM perhaps holds the only road network data set.
IMO, OSM’s strength isn’t hosting existing data sets anyways. It’s our ability to thoughtfully use all available data and human survey and inspection to create a best of breed map/database.
Only true if OSM uses available data…
Anyway Divisions will be the first Overture theme to drop OSM data, just because the others are harder, doesn’t mean they wont do it.
A dramatic change like that would cut both ways. Overture would be able to make some bolder claims about the utility and novelty of their dataset, while we – and the remaining redistributors of OSM data – would be able to more clearly differentiate our offering from theirs. Overture’s initial value proposition was essentially “value-added OSM”, which they clearly want to grow out of to some extent. Judging from some of the reactions in this thread, the discomfort is mutual.
Google initially used TIGER and at one point in time they published a paper on how they cleaned it up. The common ancestry is the reason that there are now and then errors that a present in both google and OSM. But I assume these are rarely turning up these days.
Given the nature of TIGER it is kind of hard to believe that they would still use it for anything, particularly geometry.
The rather surprising thing is that Marc said the quiet bit out loud and that in a piece that they essentially controlled.
That naturally wont stop the claquers clapping. At least it is something that we can point to for orgs that never bothered to become OSMF members, but are now paying more to enable C level grift
The other thing is that the Linux Foundations Overture Maps toolchain continues to be proprietary
Let that sink in.
And that’s just a financial negotiation between Facebook et al and TomTom. It’s not like the road network data doesn’t exist.
But is it being maintained?