[OSM-talk-be] Tagging proposal for cycling highways (Fietssnelwegen)

Mon Dec 23 22:40:08 UTC 2019

I agree with the remarks of Stijn. Only the parts of the 
"Fietssnelwegen" that are realized and “Befietsbaar” on the website of 
Fietssnelwegen and/or marked in the field as such, should be on OSM as 
cycle route.
During the past 2 years I suffered several times from the unreliable 
information on OSM as a user of OSM based bike route planners. Planned 
cycle highways were put on the map as realized and existing. A bike 
routeplanner makes a route with preference to cycle routes that are on 
OSM. I supposed to follow a cycle highway but landed on a single track 
path of 30 cm wide with surface of soft sand that I had to walk. On 
another spot I was following a paved footway and had to squeeze my 
brakes at once because the paved footway went over in a stairs downwards 
where a bridge will be build in the future. Luckily it was in daylight 
and feasible; users of cycle highways are supposed to take these routes 
before and after work when it is dark.The proposed routes on OSM are 
dangerous.

I have given that cycle highway relation the state proposed=yes that 
makes that they are not taken in account on bike routeplanners and on 
https://cycling.waymarkedtrails.org 
<https://cycling.waymarkedtrails.org/> (those proposed relations are 
visible on the Bike Map layer on OSM cycle map layer 
<https://www.openstreetmap.org/relation/1244996#map=15/50.7919/5.4333&layers=C> 
). There was a fixme or incomplete remark on those relations of planned 
cycle highways but those doesn’t make that they are neglected by 
routeplanners.

I have put the proposed state on other cycle highways that were mapped 
as going through fences over private industrial premises and others 
where biking was not permitted or where even was no path at all.

I have deleted parts of cycle highways in the route relation where bike 
riding wasn’t possible as for example on railway bridge 
<https://www.openstreetmap.org/way/242291877#map=17/51.07556/5.21979&layers=C> 
where the bridge wasn’t ready a few months back (maybe it is meanwhile, 
but I wasn’t there recently).

A few years back I have mapped _parts_ of cycle highways that where 
ready and marked and put on the website as “Befietsbaar” in a route 
relation but I had to notice that parts that weren’t ready were added to 
those relations.

I also don’t like the “alternative cycle highways” because they only 
exist in the head of one person and their quality is (in a lot of cases) 
very poor and dangerous. Example: 
https://www.openstreetmap.org/way/17298358 If you take this path riding 
on modal electric bike style downwards from the embankment of the canal 
over a small unpaved path to a narrow bridge over a ditch, you are 
death. And that should be highway for bikes.

I propose to *delete all what is “**alternatief Fietssnelweg” *because 
they are non existing and they make OSM unreliable because those routes 
are put as preferred by routeplanners.

For the F Fietswegen I propose to *delete the parts that are not ready* 
from the route relations and leave the parts that are ready and 
“Befietsbaar” as on the on Fietssnelwegen website (putting the 
“proposed” status to a complete F relation isn’t a solution any more 
because parts of them are released as “Befietsbaar”).

Regards,

Eebie

Op 23/12/19 om 21:10 schreef Stijn Rombauts via Talk-be:
> Hi,
>
> I don't understand why nobody else objects to the 'alternatives'. 
> They're just somebody's personal inventions, but they do not exist. If 
> we allow Jo's alternatives, then we have to allow anybody's 
> alternatives, suggestions , etc. for cycle highways or any other kind 
> of hiking, cycle, ... routes. E.g. the cycle highway between Diest and 
> Hasselt has been deleted: can I add to OSM a good alternative that I 
> use daily? I hope the aswer is no. I don't mind that somebody suggests 
> on some website alternatives for the cycle highways which do not yet 
> exist. It's even a very good idea, but please keep them out of the OSM 
> database.
> In my opinion, only those parts which are already waymarked should be 
> in OSM as cycle highways (and shown on e.g. 
> https://cycling.waymarkedtrails.org). The fact that there is a road or 
> a cycle path which might be turned into a cycle highway, doesn't mean 
> that there is a cycle highway. So, all the rest: state=proposed. [As 
> it is already difficult enough to keep OSM a bit up to date, adding 
> things which might be realised in some distant future seems to me a 
> bit of a waste of time. But that's just my opinion. Anyone is free to 
> do so.]
>
> Regards,
>
> StijnRR
>
> Op dinsdag 10 december 2019 16:23:51 CET schreef Jo <winfixit at gmail.com>:
>
>
> Hi Pieter,
>
> You are right, that is an odd way of tagging them. cycle_highway seems 
> better indeed. I don't know who started doing it that way, I simply 
> continued the practice, without giving it enough thought.
>
> Most of these cycle highways can't be cycled from beginning to end, 
> they continue over large distances (for bicycles). This means they are 
> all tagged with state=proposed. Some of them are mostly done though, 
> like F1 or F3, but the parts that are missing from them will take 
> several years to complete. Do we want to keep them with state=proposed?
>
> What I started doing is to also map alternatives that can be cycled 
> from start to end today. I recently learned this is not really 
> appreciated by some official instances. They don't control what we do, 
> so it's not extremely important, but still maybe something to keep 
> in mind.
>
> One thing I was considering to do, is to divide them in subrelations. 
> Such that the parts that are finished would go into both the 
> 'official' relation and into the alternative one. If you would like, 
> I'll do this for F3, to show what I mean.
>
> Then there is also sometimes  a difference between what is shown on 
> fietsnelwegen.be <http://fietsnelwegen.be> and what is actually 
> visible in the field. I'm thinking about the situation in Veltem, 
> where F3 has a leg on the southern side marked in the field, but it is 
> actually meant to go through the center of Veltem, north of the 
> railway it generally follows.
>
> Most cycle highways are not yet visible in the field. The signs 
> aren't  placed yet. For example F203 from Sterrebeek to 
> Sint-Stevens-Woluwe. It passes through Kraainem over 2 cycleways of 
> 50cm, with no separation to motorized traffic that is allowed to go at 
> 70km/h there. Then it goes through the center with lots of crossings. 
> This is a bit odd, as there is the possibility to pass through 
> Molenstraat, wich is a lot safer and has a far better experience for 
> the cyclist.
>
> The alternative route relations I was creating, are meant to disappear 
> after a few years, but that point, I might be tempted to keep it, even 
> when the official instances decide to keep the less suitable itinerary.
>
> One general problem with the cycle highways, today, is that it's next 
> to impossible to apply 'ground truth'  to them, except if we would 
> only map the parts that are actually already finished and marked in 
> the field.
>
> Those are my thoughts on the subject. If I find some more time, I 
> might continue mapping the official ones, with the projected parts, 
> like I did it here: https://www.openstreetmap.org/way/691027464/history
>
> But for longer stretches. I have no idea if they are planning to add 
> those dedicated cycleways in the next 2 years, or in the next 15 years 
> though.
>
> For the ones that I audited over the past year, there are many 
> pictures on Mapillary.
>
> Polyglot
>
> On Mon, Dec 9, 2019 at 10:53 PM Pieter Vander Vennet 
> <pietervdvn at posteo.net <mailto:pietervdvn at posteo.net>> wrote:
>
>     Hello everyone,
>
>     As we (Anyways BVBA) are making a route planner which takes
>     'Fietssnelwegen' into account, we would like to have some uniform
>     tagging into place for this.
>
>     Some of them are already tagged with `cycle_network=Fietssnelweg`,
>     which
>     sounds very Flemish.
>
>     I'm going ahead with adding them to other existing fietssnelwegen, but
>     would like to document them on the wiki and to have some more thought
>     put into them. First of all, the dutch term is something very
>     inconsistent with the rest of OSM - perhaps "cycle_highway" is a
>     better
>     fit. Secondly, maybe we should prefix them with "BE:". Thirdly,
>     OSM tags
>     are mainly written in lowercase, which this tag is not.
>
>     Any more thoughts on tagging? I'm especially looking looking
>     forward to
>     input from polyglot who is very familiar with them.
>
>     -- 
>     Met vriendelijke groeten,
>     Pieter Vander Vennet
>
>     _______________________________________________
>     Talk-be mailing list
>     Talk-be at openstreetmap.org <mailto:Talk-be at openstreetmap.org>
>     https://lists.openstreetmap.org/listinfo/talk-be
>
> _______________________________________________
> Talk-be mailing list
> Talk-be at openstreetmap.org <mailto:Talk-be at openstreetmap.org>
> https://lists.openstreetmap.org/listinfo/talk-be
>
> _______________________________________________
> Talk-be mailing list
> Talk-be at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-be

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-be/attachments/20191223/d6dc381f/attachment.htm>