"We parsed OSM’s complicated building and building:part tags to refashion our building features from the ground up. "
What could it mean?
1 Like
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?
1 Like
If somebody dreams of sticking knifes in to OSM, this might literally be the call for you https://twitter.com/pwramsey/status/1630597756355231744
Just reading the Minutes of the February Board Meeting & noticed: Board/Minutes/2023-02-09 - OpenStreetMap Foundation
“The board discussed their 2023-02-07 meeting with the Overture Steering Committee. Discussion during the mid-month chat not minuted after request.”
Sorry, but that’s definitely not a good look, as it appears that there’s something that you’re trying / need to hide?
1 Like
I suspect that this is to avoid leaking anything of importance to overture.
4 Likes
Hopefully it’s that, & not vice versa 
1 Like
I image there’d be NDAs involved for anyone from OvertureMaps to talk about it since the software is still pre-release.
No NDA were signed or asked to be signed, as far as I know.
I definitely have not signed any Overture NDA (or anything similar).
11 Likes
April 12, 2023
Overture Maps Foundation Update – A Look at Progress, Possibilities
https://overturemaps.org/april-overture-release/
6 Likes
I noticed:
" Commercial-Grade Quality Checks. The core dataset in this release combines the contributions of the global OpenStreetMap community with quality and consistency checks to create a free, stable, and easy-to-use map. We put the data through multiple validation checks to find as many of these problems as we can: broken data, bad words, vandalism, etc. Overture Maps will expand on these quality checks to remove as many quality problems as possible."
Wondering what they’re going to do when something is actually named with a “bad word”? Not show it at all, or rename it? 
At a guess, they would build up a list of valid “bad words” along with their location. This way they can let genuine cases through whilst still removing vandalism.
Facebook has a project called “Daylight Map” which I suspect this is somewhat related to. In January I had a look at a couple of English counties from that and from an OSM extract from around the date that the Daylight data was extracted. Things missing from the Daylight map included:
- Admin boundaries (they’re a separate download)
- Things without meaningful OSM tags (e.g. just a “name”).
and that’s it. I didn’t see any censorship of “valid bad words” in the Daylight data, such as this artwork - it was in both datasets. I didn’t see any actual cleaning of “bad words” that shouldn’t have been there, but to be honest I wouldn’t have expected there to have been any in the source data - vandalism of this sort is usually fixed pretty quickly in OSM.
Separately various people (including, but not limited to, people from Facebook) look for and remove “bad words” that aren’t valid from OSM itself. That doesn’t always work 100% (see for example here), but it’s not a major source of complaints to the DWG** about Facebook. There’s a writeup of Facebook’s process here.
** for completeness, both me and @Fizzie41 are DWG members, and people from the DWG regularly remove “Finbarr Saunders”*** level vandalism
*** a comic strip in Viz.Comic - “Finbarr Saunders and his Double Entendres”. Google it if you must.
1 Like
Yeah, I was wondering how that famous English town of Scun… will go?
& from glancing through the list of deleted things, was also wondering if they ever refer users to the DWG for blocking? You ever had one come through this channel?
There was a MapRoulette challenge about Profanity detected by OSMCha. In the end 1.4 % of flagged issues were fixed, 2.6 % were “already fixed” by other users outside the challenge, and 96 % were false positives. The few cases I looked at were “bad words” in English that were fine in the local language or were the actual, correct names of places.
1 Like
With regard to Maproulette, it seems like about the last tool that I’d choose for this task, since remote editors will lack on the ground knowledge. See also here, which is a discussion about Maproulette and the difficulty of finding out what has been done wth it.
Am I misunderstanding something?
https://overturemaps.org/april-overture-release/
“Today we are making a pre-release data set (2023-04-02-alpha) available that has incorporated some of the new features that we plan to build in Overture.”
“we’re leaning heavily on OSM data… The core dataset in this release combines the contributions of the global OpenStreetMap community”
https://overturemaps.org/download/overture-april-alpha-release-notes/
“it is open data and available under the CDLA Permissive V2.0 and ODbL licenses”
You can’t take OSM data and redistribute it as CDLA Permissive. Are they saying that their download is a Collective Database where (somehow) some “layers”, presumably identified by tags, are CDLA-Permissive and the rest is ODbL? Or am I entirely missing the point (always possible)?
5 Likes
I suspect that now that this has been pointed out someone will quickly apologise for the “mistake”. 
FWIW Facebook’s Daylight map data** is ODBL licenced (the statement is “Daylight will support the open licenses of upstream data projects like OpenStreetMap”).
** which it sounds like this basically is, apart from e.g. some buildings in San José
3 Likes
While technically the actual product might not contravene the ODbL, they are clearly not making the licenses of the constituent parts clear. The release notes -do- contain a licence link right at the end “Attribution” but I don’t think that is anywhere near obvious enough, and in particular it still doesn’t really clarify the situation for a potential user.
IMHO I would send the Linux Foundation in Brussels a letter of the appropriate niceness level.
10 Likes