Talk:Proposed features/incline up down

From OpenStreetMap Wiki
Jump to: navigation, search

That's a good idea. I can speak for myself, I have no idea how could I measure the degree on inclination of a street, but it would be nice to map if a road is inclining or not. --Nighto 14:46, 20 August 2009 (UTC)

A feasible estimation technique, albeit slow if done with an aim for precision, is to
  1. Measure the distance from the floor to your eyes (h)
  2. Stand at the bottom of the steepest section
  3. Estimate the point where the horizon meets the incline; the ground point which is at the level of your eyes
  4. Walk there; measure distance (d) from the gps log (or estimate, or measure with "one meter strides")
  5. now h / d * 100 = incline percentage.
Or if there's a building in the incline, take a picture perpendicular to the incline direction; choose a reference distance (one floor is often 2.8 to 3.5 meters), count pixels in both directions and then do some divisions; you get the same height/distance. Alv 16:54, 20 August 2009 (UTC)
The % inclination is how many % you go opp according to a distance travelled, i.e. if you go 10 meters up on 100 meters distance, the incline is 10%. --Skippern 00:55, 21 August 2009 (UTC)

Direction of steps

I personally use the direction of way for highway=steps in the (architectural) standard way that the way point upwards, i.e. the bottom of the steps is the beginning of the steps-way and the top the end. If we could agree to this general definition a separate incline-tag would not be neeeded. -- Dieterdreist 14:48, 20 August 2009 (UTC)

This has been discussed 18 months ago, and we didn't agree on a common definition since then. The biggest problem (besides the thousands of existing steps that would need to be reversed) is that there is no way to tell whether the person drawing the steps was aware of the convention. What I do is to draw all my steps upward and add an incline=up, thus preventing any possible misinterpretations.
Also, the "steps always point upwards" convention might not be an ideal solution for things like escalators, where most people (including me) will probably consider the moving direction the "natural" way direction. --Tordanik 15:37, 20 August 2009 (UTC)
Also, there are other factors that may force the direction of the way to be downhill - e.g. if a way is one-way. Incline=* is useful for this and is also more explicit in general. --Waldo000000 21:29, 20 August 2009 (UTC)
The % value can be calculated on stairs just as easy as on roads, maybe easier, though up/down values might be easier. I would strongly discurrage this on roads though. --Skippern 00:55, 21 August 2009 (UTC)
For me the only question is if we need a seperate key for up/down. Or if we stick with one key. up/down is better than a completely wrong estimate - but both is better than not knowing whether a way goes up or down at all. There are keys like mtb:scale:uphill and mtb:scale that can only be used for autorouting if the up/down direction is known. And public domain SRTM with it many holes and mistakes is definitely not the solution for this. incline=up and incline=down are commonly used already. See http://osmdoc.com/en/tag/incline/#values --Extremecarver 19:52, 11 September 2009 (UTC)

Proper documentation of current use

I think it is good idea for several reasons. Firstly, this suggestion is actually just properly documenting current practice, and as such cannot be a bad thing. If we do have a better proposal later that will solve all (most) of the problems of this one without introducing new problems, than we can add recommendation to use that instead of up/down. We actually do that already - *if* you do know exact percentage of incline. However, if you don't, and can't measure it relatively accurately (for example: mapping from out of copyright detailed maps, or from donated data, or from memory, ...), I think it is better to just indicate that it is there, then to try to make up some percentage which might possibly be very wrong. --mnalis 11:45, 6 October 2009 (UTC)

Other uses

And even such vague data as just indicating there is visible incline is exceptionally useful in some circumstances when you mostly don't have it officially marked. For example, while motorways will usually have the signs if the incline is big enough (so you can enter exact number), most footways won't; and in some situations (like doing routing for wheelchairs) even the vague the incline tag might be acting more like a barrier tag, suggesting that other routes are better alternatives. This is just one example, but it has other uses also, like useful information for cyclers or walkers, so please don't suggest barrier=wheelchair instead --mnalis 11:45, 6 October 2009 (UTC)

Comment Zkir

I agree with Skippern this tag is meaningless. Absolutely horizontal roads are very rare, practically every road has incline up or down. -- Zkir 14:53, 9 October 2009 (UTC) Copied from voting section

There are many features where the general direction of incline is meaningful (such as steps or ways tagged with mtb:scale:uphill in addition to this - see Extremecarver's comment). I expect that mappers will be able to identify where the tag is meaningful, and will only use the tag where it makes sense. Do you really consider the tag meaningless for these examples, too? --Tordanik 15:56, 9 October 2009 (UTC)

direction tag

As skyper has pointed out, the tag "direction=up/down" has been used to mark general direction of incline on steps, so for steps it is an alternative for incline=up/down. This tag has been added to the highway=steps page some months ago - the edit wasn't discussed, though, and apparently wasn't announced to either mailing list or forum, so I managed to miss that edit.
OSMDoc is rather inconlusive, it shows 176+85 uses of incline=up/down and 120+71 uses of direction=up/down (there are over ten thousand uses of direction=clockwise, which hasn't anything to do with steps or incline). Personally, I think that "incline" is more specific - direction could be used for many other things, for example I'm using a conveyor_direction=forward/backward tag in my proposal for escalators and the like - so I continue to suggest using incline=up/down for this case. --Tordanik 23:32, 16 October 2009 (UTC)

disagree with this vote

I wonder how this vote ended. There were 3 votes on the 17th of October from user with no user-page at all. I think there is more discussion needed !! For myself I won`t use this tag and rather will tag incline:estimated=*. I still think it is important to tag the incline but I think especially for footpathes,skitracks etc. an estimated incline gives you more information and also shows the direction --skyper 13:22, 19 October 2009 (UTC)

Yes, this looks suspicious. GRappa and Tib are names of users whose OSM accounts have existed for some time, though, so I won't go as far as claiming that that they are vote puppets. Still, I don't deny that it is a strange occurrence. Does anyone have an explanation for this? Was there some message/post/chat mention I'm not aware of?
Nevertheless, the outcome of the vote was clear - even without those 3 votes, it would still be a 11:3 result, so I don't see a reason to question the result on the grounds of a formal quorum. Of course, you are free to not use these values - and, after all, it is explicitly recommended to overwrite them with even imprecise incline percentages. --Tordanik 22:07, 19 October 2009 (UTC)
I've contacted one of the owners of the OSM accounts sharing the name with the "red" votes. It turns out that a supporter of the proposal has informed some mappers who usually don't contribute to the wiki (but for whose OSM contributions the tag would be useful) about the proposal so they could still take part in the vote in time. Thus, the votes are legitimate, and I apologize for suspecting manipulation. --Tordanik 17:58, 20 October 2009 (UTC)