orientation of gpx sections inverted

I like and use waymarked trails very often to plan my own trips, either cycling or hiking or or …
But also often, it makes me a lot of work.
I select a waymarked trail and download the gpx-file. Frequently I get a track with a lot of sections which have inverted orientations. As a result I get a track which looks like a net rather than a path. It’s a challange for me to get the sections cut and collected in the right order/orientation.
Example: A waymarked trail runs from A via B via C via D to E.
The orientation of the section A-B is left to right, means AB
the orientation of the section B-C is right to left, means CB
the orientation of the section C-E is left to right, means CE
and so on.
Can anybody of you tell me the reason and how to prevend it?
This could help me a lot.
Thank You


Please give a concrete example.

This could be a problem with waymarkedtrails (for which contact them), or it could be one of two types of problem with the route relation that it is using:

Firstly route relations often get mis-sorted;

Secondly, it is possible that one directional route segments are specified in the route in the wrong direction.

As an interim measure you could plot your route using gpsies.com as it enables you to have waymarkedtrails as a background map. It has an option to download a gpx of your track, etc.

I wouldn’t assume that any relation that you download from OSM (including via Waymarked Trails, I guess) is sorted. Many or most edits to OSM relations won’t make any attempt to sort data.

If you want sorted data, you’ll either have to sort it yourself or find a site that sorts it for you as part of the download process. Many of the sites providing access to this data are based on open code, so you may even be able to make the change yourself and suggest to the site owner that they incorporate it.

Weserradweg D9 comes back as not found on both OSM itself and waymarkedtrails.

Could you click on it and give the relation number from the top of the information box?

As people say it is probably mis-sorted, although, as editors go, JOSM makes a good attempt to retain sorting, and provides tools that help to correct it.

Actually, based on a search by source code, there may be a sorting tool in JOSM.

Do note, though, that routes with alternatives may have no unique sort order.

The wiki page on routes has a section recommending that mappers order their routes, and pointing out some of the issues you mention that arise from unordered routes: https://wiki.openstreetmap.org/wiki/Relation:route#Order_matters.

However I suspect most people who map hiking and cycling trails never read that, as they are more likely to start with the more specific pages on hiking/cycling routes, which make no reference at all to ordering.

I believe this is the specific relation you referred to:

In addition to the points already made, this is a relation of type “superroute” whose members are themselves relations of type “route”. Superroute relations are not very widely used (it looks like there are only around 400 in the world) and have limited support in editors. In fact if I remember rightly the JOSM validator complains that it is an unrecognised relation type. So even if a mapper wanted to put the members of the relation in order, it’s not necessarily an easy thing to do - especially for a route like this that is more than 1,000 km long.

Looking at some of the sub-routes within that relation shows examples of hadw’s point that routes with alternatives may have no unique sort order. Look at the section around Bad Mergentheim (https://cycling.waymarkedtrails.org/#route?id=2604308&map=13!49.4997!9.8002). There appears to be a tangle of alternative routes. How would you order those? It looks like the mappers have made a reasonable attempt to put the “main” route in order, and then added the short alternatives at the end… which of course makes a mess of the elevation profile. But what else could they do?

In summary: many mappers don’t try to order their routes, either because it doesn’t occur to them or because they believe it should be handled by the applications that use the data, and those of us who do try often find it is not as easy as it might seem at first glance.