[EuroVelo] EV17

As discussed here, this topic is dedicated to EV17 and its mapping in OSM.

What makes it interesting is its structure. The following image shows the GPX provided by EuroVelo and:

  • its two branches (stages 24-25 and 26-28 in the GPX file)
  • the way the French and Swiss parts are sequenced

How would you organize and tag it?

1 Like

I did check the GPX file with the current mapping of EV17.
An idea of the segments and stages could be:

EV17 segments and stages

  1. Andermatt - Le Bouveret
    2a Le Bouveret - Genève (north side of Lac Leman)
    2b Le Bouveret - Genève (south side of Lac Leman)
  2. Genève - Lyon
  3. Lyon - Beaucaire
    5a Beaucaire - Port-Saint-Louis-du-Rhône
    5b Beaucaire - Sète

Segment 3 could be split at the border between France and Switzerland
The stages of the GPX file could be used with only a few changes at the current mapping:

EV17 segments and stages of the segments

  1. Andermatt - Le Bouveret
  • Andermatt - Oberwald
  • Oberwald - Brig
  • Brig - Sierre
  • Sierre - Martigny
  • Martigny - Le Bouveret (current mapping: Martigny - Montreux)

2a Le Bouveret - Genève (north side of Lac Leman)

  • Le Bouveret - Morges (current mapping: Montreux - Morges)
  • Morges - Genève

2b Le Bouveret - Genève (south side of Lac Leman)

  • Le Bouveret - Thonon (part Le Bouveret - Saint-Gingolph currently not mapped as EV17)
  • Thonon - Genève
  1. Genève - Lyon
  • Genève - Vulbens
  • Vulbens - Seyssel
  • Seyssel - Belley
  • Belley - La Balme-les-Grottes (currently split at Groslée)
  • La Balme-les-Grottes - Jons
  • Jons - Lyon
  1. Lyon - Beaucaire
  • Lyon - Saint-Romain-en-Gal
  • Saint-Romain-en-Gal - Sablons
  • Sablons - Valence (currently split at Tournon-sur-Rhône and having an alternative route)
  • Valence - Châteauneuf-du-Rhône (currently split at Le-Puzin/Cruas)
  • Châteauneuf-du-Rhône - Pont-Saint-Esprit (using Lapalud as to instead of Pont-Saint-Esprit)
  • Pont-Saint-Esprit - Caderousse (using Lapalud as from instead of Pont-Saint-Esprit)
  • Caderousse - Avignon
  • Avignon - Beaucaire

5a Beaucaire - Port-Saint-Louis-du-Rhône

  • Beaucaire - Arles
  • Arles - Port-Saint-Louis-du-Rhône

5b Beaucaire - Sète

  • Beaucaire - Saint-Gilles
  • Saint-Gilles - Aigues-Mortes
  • Aigues-Mortes - Palavas-les-Flots
  • Palavas-les-Flots - Sète

For the south side of the Lac Leman and then to the Mediterranean the current mapping is using the ViaRhôna relation. We could create EV17-stages from the stages of ViaRhôna because we shouldn’t change the ViaRhôna part relations.

So the one really change would be to split the stage Martigny - Montreux at Le Bouveret and adding the part between Le Bouveret and Montreux as first part of Montreux - Morges.

Would it make sense to split the transnational stages at the border:

  • Le Bouveret - Thonon at Saint-Gingolph (no work, because currently not existing)
  • Thonon - Genève at Hermance
  • Genève - Vulbens near Chancy (current mapping of Route du Rhône Étape 8)

Same for the segment Genève - Lyon.

Issue missing part south of the Lac Lèman between Saint-Gingolph and Le Bouveret (Bridge over the Rhône).

The GPX track of this part seems nearly identical to tht part of the Tour du Léman

Edit (of this topic):
Adding eurovelo to the tags of this topic

With your proposal in mind, I looked into the current structure of the relation as if I was going to change it. Let me try and summarize the decisions that we need to make based on this observation:

  1. do we keep current “national” relations? Your proposal seems to reply no. Truly, nothing prevents us from keeping them separately (here, not in the structure of EV17, but as two independent super-relations that have members in common with EV17).

  2. what do we do with the names (and other related tags) of segments and/or stages when they already have a name that has a particular meaning relative to the national route (e.g. Rhone Route Etappe 1). I do not see an explicit reply in your proposal, but that is where Nadjita’s proposal could help and/or be tested

  3. what do we do when the national segments/stages do not match the european segments/stages? You propose to keep the national segmentation (Via Rhona).

  4. how do we handle branches such as 2a/2b and 5a/5b, with route continuity in mind? Do we use the alternative role? Do we use a master_route like for buses? Another option? I do not see a reply to this question in your proposal, and I believe there is room for improvement in the current practices.

  5. do we split sections where they cross borders? I understand that you propose a flexible solution: split when it is useful (here, in segment 3 to preserve Via Rhona I guess). Other options could be to say no because there are no national relations in the structure anymore, or yes because we need different operator tag values and we can always make stages out of two sub-relations (one for the part of the stage in each of the two countries)

Remember, we want to test a new tagging method, currently only a private proposal draft.
I would change not more than is needed. So I propose to keep both current existing relations. It is a national swiss one (Etape 1 to 8) and and international one.

One of the new tagging policy of @Nadjita’s draft is not using names for stages or segments that have no name, so I propose to create new relations with the new tagging method that contain most time one element of the existing ones, sometimes two.

Is ViaRhôna an existing independent bicycle route or is it just another name for the EV17? This route seems at least completely identical with that very great part of the EV17. I think: do not change the existing one. Create one new stage from Le Bouveret to Thonon (with using the first stage of ViaRhôna), otherwise using just one ViaRhôna stage for the new stage, sometimes two. It would not too much work and not disturbing existing mapping.

I would test them as main segments, not as alternative. It would be easy to retagging that, if it doesn’t work out.

I’m open for both: splitting the stages and splitting the segments - or using multinational ones like ViaRhôna already does at the stage Genève - Vulbens. But I have to say that for the route south of the Lac Léman I would prefer one single segment instead of three.

If we decide for splitting at borders, we have to:

  • create the new Le Bouveret - Thonon stage,
  • either split the existing Thonon - Genéve stage or simply two new ones
  • create a new one Chancy - Vulbens

Seems not too much work either.

I think it would be better if all stages are EV17 stages.
Therefore the proposal to create a complete new set of stages without modifying the existing ones, maybe with the exception of Martigny - Montreux and Montreux - Morges, but we can even create two new ones Martigny - Le Bouveret and Le Bouveret - Morges, because we may need them later, if our tagging scheme will be the accepted one.

We will have a double mapping but I don’t think that’s too bad for a test in production and there is less potential of conflicts. If we would remove the name tag from the existing ones that would produce some conflicts for sure.

What about Knooppuntnet. From your national compilation I deduce that you can decide which relations will be used. Is that correct? If yes, we could use the new stages. But we should document the tagging scheme for stages and segments. Maybe here, maybe in the EuroVelo tagging topic.

I’ve seen that there are some ViaRhôna parts that could be used for alternative (at Valence) and approach too.

I’ll give a try for the tagging:

Segment Andermatt - Le Bouveret

name=EuroVelo 17 # or should it be Rhône Cycle Route or even EuroVelo 17 Rhône Cycle Route?
to=Le Bouveret
description=? no real idea
ref=EV17 # is that right? Or should it only be 17 and using a label too?
network=icn # or is this only ncn?
operator=Veloland Schweiz #maybe only for stages?

Stage 1 Andermatt - Oberwald

description:en=The first stage leads from Andermatt to Oberwald
name=EuroVelo 17 # see my questions above
network=icn # or is it indeed rcn?
operator=Veloland Schweiz
# ...

Maybe @Nadjita could comment and correct this first try according to the proposal draft.



We will not have any conflict because we discuss that at European level.

If we try a new tagging scheme that is only a draft I see the potential of conflicts.
If ViaRhôna is a separate route of its own instead of another name for EV 17 than we need new EV17 relations for the greatest part and it’s only the 7 stages of the Rhône route. Two persons in favour of this new scheme are not enough to do this massive change without an agreement at least here or at the wiki page.

I’m not brave enough to do that …

this isn’t a name :wink:

This is ref, so this tag (at least here isn’t useful)

So we shouldn’t use it.

Editors (JOSM, iD…) but also tools (KPN, WMT…) should display a label with the following expression : ref + "/" + from + " — " + to (this is an example. See Display Name from Nadjita proposal )

Right, but for the moment we just discuss this.
English isn’t my mother tong, so this is easier for me if we could describe only the result.
(but ok, to describe a temporary step in a note)


I’m using the examples of the draft: https://wiki.openstreetmap.org/wiki/User:Nadjita/Routes#When_to_use_ref=*_and_when_to_use_label=*?
The one of Burford Station with the E2 hiking route. But I’m asking here to find out what would be the appropriate tagging.

I was surprised myself seeing that example.

Ok :grinning:

Some part of his proposal seem “good”, other less.
In my mind we could use the obvious and “good” parts now, a leave the “bad” for later :sweat_smile:

1 Like

What about that:

  • removed: name, short_name
  • added: ref, description:en, cycle_network
  • corrected: type
  • confirmed: operator (by current use)
to=Le Bouveret
description:en=The alpine part. From the heights of the Alps to the Lake Geneva
network=icn # or is this only ncn?
type=superroute #route was wrong
operator=Veloland Schweiz #that's currently used for the superroute from Andermatt to the French border

This does not work if the section belongs to another superroute at the same time.

This is new to me, we don’t have the equivalent for hiking routes. Actually I have seen it a few times when splitting EVs, but with very inconsistent values and generally this is the sign of a non-standard / unused tag. Is it used anywhere?

I copied it from the EV example of Nadjita. I made an overpass-turbo query and it is used a great lot. We could include it in the new EV-tagging.

I did not mean “used in tags”, I meant “used in applications”. It does not feel very well defined to me, a bit like the infamous “symbol” tag.

For stages this is true, but whole segments, I just don’t know. But it’s not impossible.

Maybe ref=EV6;EV15

Looking at a ViaRhôna stage that is identical to a EV stage. I wouldn’t modifiy the ViaRhôna stage and create a stage for EV17 with the EV-tagging and the ViaRhôna relation as single member.

Another option could be to avoid using ref in segments and stages, and leave it to superrelations.

We could see in WMT that “EV1” is displayed twice (in ref and in osmc:symbol) for each stage.

If we remove “EV1” from ref in stages we should logically remove it from osmc:symbol ?

Also if we search “EV1” in WMT, it display 21 stages and not the main relation itself.

If we keep it from osmc:symbol, we should have two values for this tag in the EV6 / EV15 example ?

No software support multi values yet for osmc:symbol key :crazy_face:

Either a name or a ref is needed to have a display name, see

If using a name (for stages or segments) no stage information should be included. If I understand @pyrog correctly we should only tag the stages and segments with ref but not with name.