Thank you that really interesting. One of the advantage of Sophox is the integration of OSM Data Items. I hope it could be easily integrated into QLever and even if possible on your demo instance. I may try myself to run an QLever OSM Data Items instance. It is the database I want the more query on.

Pour les données OSM ça peut marcher avec des nodes des ways ou relations, ça se configure. Pour les données OpenData, oui, que des points.

Il existe déjà des comparaisons de ce type d’objet dans Osmose-QA. http://osmose.openstreetmap.fr/fr/issues/open?item=8150

Il faut créer ce genre de configuration https://github.com/osm-fr/osmose-backend/blob/master/analysers/analyser_merge_cycle_parking_FR_bm.py

Ça peut traiter tous les cas de différences que tu signales https://github.com/osm-fr/osmose-backend/blob/master/doc/4-Merge.md

Add a new patch to use only logo from commons.wikimedia.org, not from facebook not twitter https://github.com/frodrigo/iD/commits/compliance_v2.16.0

I switched http://id.openstreetmap.fr/ to v2.16.0.

My idea is not to fork iD and even less to maintain a fork. But rather show the list of controversial decisions could be addresses with patchs. The number of, and the patchs are not large. Writing the patchs and maintained can be done. This approach could probably be applied to brand logos and so on. This changes are mostly configuration, or configuration-like part of iD, not real software code.

Because I obliviously care about the OSM Data. I also agree with the content of this diary. We should no more deploy the default iD on osm.org. I aim to show a set of patchs can be done and maintained.

Few week ago we also add a new check in Osmoss-QA about this, it reveals lot of (in)interesting thing:

http://osmose.openstreetmap.fr/en/errors/?item=5070&class=50703

And the slipy smiley map : http://osmose.openstreetmap.fr/en/map/#item=5070&class=50703&zoom=9&lat=49.402&lon=8.364&layer=Mapnik&overlays=FFFFFFFFFFFFFFFFFFFFT There even one in the holly city of Kalsruhe

Mixed language also come this the same check: http://osmose.openstreetmap.fr/en/errors/?item=5070&class=50701

You can look at this for Mapillary, but in changeset comment, not source : http://resultmaps.neis-one.org/osm-changesets?comment=mapillary

In fact Osmose QA already check for this.

http://osmose.openstreetmap.fr/map/#layer=Mapnik&overlays=FFFFFFFFFFFFFFFFFFFT&item=1120&level=1%2C2%2C3

http://osmose.openstreetmap.fr/errors/?item=1120

Source code is here : https://gitorious.org/osmose/backend/source/f80d9181a2e21e1be78f607e99bd1cc791413740:analysers/analyser_osmosis_broken_highway_level_continuity.py

This errors set is also available as a Maproulette challenge.