Hi Ilya,

I came across your diary and felt sad reading it, because of your disappointment and because I feel I’m partly cause of it. I also feel sad because you seem to suggest that I influenced french mappers to vote against the proposal. By no means I would consider doing this ! I do know two persons who opposed, they did not need my influence : both have been involved in mapping subway stations before I even started doing it. The two persons that actually work with me on this project have not taken part to the vote.

I also want to clarify that we (Carto’Cité) are not actually using the data, we just look after the data on behalf of SNCF (the french railway company). And to do so we actually developed a piece of code that analyzes Overpass Augmented diffs. We called it OSMADA and published with an free software Licence : https://github.com/Cartocite/osmada.

The one thing I feel guilty about is not to have spotted your proposal early enough to discuss it prior to the vote. I can only apologize for that now. Maybe the proposal tries to tackle too many issues at once, which makes it hard to understand the full picture. There are bits I have no problem with, such as the stop_area_group relation type, others that make a lot of sense even though they impact data used by SNCF (avoiding these massive relations for Paris railway stations which I don’t personally like), and others that I feel are wrong, such as forcing to map stations as nodes. The latter would mean undoing a lot of work that has been done by mappers, and not just our work : mapping stations in the Greater Paris was started before I got involved in this project.

Now where do we get from here ? I understand you disappointment however I have faith in the OSM community to find a way to come to a compromise. I don’t think that leaving the proposal “not approved but widely used” would be satisfactory for anyone. Could there be a way to split out the proposal into more digestible bits ?

Maybe that situation we’re facing today was to be expected : the more people are making use of OSM the harder it gets to change things. But harder does not mean impossible, it just takes more explanation (and maybe the proposal mechanism needs to evolve, as your post title suggests).

Regards, Antoine.