Re: [OSM-talk] validated osm by linux fondation ?

2022-12-17 Thread Stephan Knauss

On 15.12.2022 21:06, Marc_marc wrote:

+ news from TtomTom that it switch to osm, that let me thing that
this new database is an osm+extarnal data+validator+schema 
"fork/downstream"


Don't worry. These companies are not stupid. And they are all having a 
long record now of contributing to OSM. Sometimes this did not went so 
smooth, but they all demonstrated interest in improving the cooperation.


Maybe think of it more as an analogy to a Linux Distribution.

We are the upstream. We have the latest and greatest data. But also 
sometimes rough edges or vandalism not yet corrected.


Certainly this raw and unvalidated data is not suitable for all 
use-cases. Sometimes you are happy with some slightly older data which 
has passed some validation. And that is then a product these companies 
can use.


Watch some of the SOTM presentations of the big companies of how they 
work with OSM data and their data validation teams. Then it will be 
obvious why you often can't directly use OSM source.


Unless you have extremely good ways of automatically merging in your 
fixes, it will always better to fix problems upstream. And to my 
understanding they are doing that.


And as long as the majority of the data originates from OSM they have a 
strong business interest to keep OSM in a good shape.


Stephan


___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk


[OSM-talk] validated osm by linux fondation ?

2022-12-15 Thread Marc_marc

Hello,

reading:
https://www.linuxfoundation.org/press/linux-foundation-announces-overture-maps-foundation-to-build-interoperable-open-map-data

+ news from TtomTom that it switch to osm, that let me thing that
this new database is an osm+extarnal data+validator+schema "fork/downstream"

it look like that don't fixing/imrpove landuse=grass 
error/counter-intuitive/mess and not fixing to so-many-error that our QA 
tools detect, we may loost the control of what's is correct and what's a 
good schema and let AFAM decide for us.


bad day... despite it 'll make the use of the osm data more easy

Regards,
Marc



___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk