mf-geoadmin3
mf-geoadmin3 copied to clipboard
geoadmin
wrong line of sight distance(path distance) versus distance on ground (Path of Distance)
The distance measurement does not give a correct result: line of sight distance(path distance) versus distance on ground (Path of Distance) does not make sense
https://s.geo.admin.ch/7a38fce666
here the distance on ground should be way higher
and since the line of sight does take into account all points: and i most cases you want to get results only from the first to last point: it would make sense to implement it like this or only show the result on the second point (like azimuth)
Taks: we only show the line of sight between first and second point (when azimut is shown) if more than 2 points are available we remove it.
Result: we only show
-
line of sight: distance between A and B in straight line, taking into account the altitude (earth curvature)
-
distance: the red line (clamped on the ground)
To do
- [ ] the eye icon and the mouseover tooltip are giving mixed infos (eye means "line of sight", mouveover says "linear distance" and returned is linear distance). We should fix that to line of sight
- [ ] remove line of sight when more tahn tow points (or: only show when azimut is shown)
- [ ] show line of sight as well in azimut label on map
- [ ] adapt help
So line of sight distance should always just be from first point to last point as on a 3D line?
I think we should first think again what this "eye" symbol means. If your read the mouseover, it is described as "Linear distance" ("Luftlinie" in DE) opposed to "path distance" ("Wegstrecke") and NOT as "line of sight" Where is the difference? "Line of sight" is the flat shortest distance from start to end and @davidoesch example makes then sense. But to me it is NOT the case. "Linear distance" means to me the flat distance of the line I draw while the Path (second measure) means the effective distance I need to WALK, thus including the ups and downs of topography.
Now my 2 cents:
- to tell the user how much ups and downs he makes is useless because, as we saw, they are not correc in the actual state. They will be interesting once the profile uses a much higher point density
- the eye icon and the mouseover tooltip are giving mixed infos (eye means "line of sight", mouveover says "linear distance" and returned is linear distance). We should fix that
- I am not sure that true line of sight between start and end could be useful. I would rather change the eye icon and make it more evident about what we are speaking
Sorry, I wrote this in a hurry. I hope you can understand what I'm saying
IMHO like azimuth: we show it only between start and first point in the tooltip
(o): distance on the paper passing though all points (flat, 2D): for the artillery distance: distance on the terrain, i.e. more or less what you have to walk (kind of 3D): for the infantery
So what are your definition for line of sight, distance, etc.
- line of sight: distance between A and B in straight line, taking into account the altitude (earth curvature)
- distance: the red line (clamped on the ground)
@AFoletti @davidoesch @gjn
@AFoletti @procrastinatio this needs a brainstorm peak to peak the https://s.geo.admin.ch/7e22b914aa it seems to be correct.
From customer feedback ID : 2019022320104849
The line distance in the KML is not the same as the line of sight as well.
might be worth investigating also how it is computed for KML. ping @procrastinatio
The above mentionned KML has 339 vertices, that means the profile service doesn't compute its stats on the original geometry, but a simplified one. Hence the differences. QGIS and feature length both give the reasonable value of 16.24 km, result from the profile (length and sight of view) are different.
So now, either we build a reasonable profile service, i.e. attributing height data to the original data. Which is what I always wanted and what is easy to understand for users. Or we continue with our hard to use/understand for the users.
And it will be fun to explain why meter in the infamous spherical-mercator is not quite a real meter, at least in Switzerland.
based on @procrastinatio and @AFoletti feedback, decision taken and adapted the to do list above in the ticket description
As an standard "outdoor" user (my2cents):
- I would recommend, when ever possible, not to modify the original KML/Drawing and I do expect the height/profile/distance service&function to return the height information based on the original input data (of course there must be a limitation given by performance&infrastructure).
- Limitation of the function/service must be visible ("known") by the users (help documentation, warning or pop-up, ...)
- It must be clear for the user if distance (mesure) is a 2D distance or 3D distance.
still open "Customer feedback ID : 2021052918371578"
https://s.geo.admin.ch/90ecec0ee6
Profil: 6.39km Distanz: 6.4km (Anstatt 1.68 KM)
as mationed above "line of sight: distance between A and B in straight line, taking into account the altitude (earth curvature)" and this is not the case here
Solution: www.schweizmobil.chhttp://www.schweizmobil.ch
De : David Oesch @.> Envoyé : lundi, 31 mai 2021 10:43 À : geoadmin/mf-geoadmin3 @.> Cc : Schütz Philippe swisstopo @.>; Comment @.> Objet : Re: [geoadmin/mf-geoadmin3] wrong line of sight distance(path distance) versus distance on ground (Path of Distance) (#4299)
still open "Customer feedback ID : 2021052918371578"
https://s.geo.admin.ch/90ecec0ee6
Profil: 6.39km Distanz: 6.4km (Anstatt 1.68 KM)
as mationed above "line of sight: distance between A and B in straight line, taking into account the altitude (earth curvature)" and this is not the case here
[image]https://user-images.githubusercontent.com/4577727/120165811-3af53580-c1fc-11eb-9f8b-a157b6f5ea72.png
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/geoadmin/mf-geoadmin3/issues/4299#issuecomment-851325369, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AC4F3EWJJTVK6GMJXTPUVD3TQNDYBANCNFSM4E5OD56Q.