OpenStreetMap
Après 3-4 mois à dégommer du mauve dans le 22, voilà une image satisfaisante.
Je me suis attelé à combler les vides dans les Côtes d’Armor à la fin de l’été 2022.
Plusieurs remarques:
I found an old mining road that’s being maintained as a trail while staying by the river, so I added that. Then I got all fiddly and added a bunch of driveways. I wish I’d taken a picture of the map BLM had on their information board at Caddis Flat Campground (added details about it) because that map had a more official trail a little further east, also leading to a mine.
Today,Imma map till I drop.. let’s get some!!
I started off my excursions around Blue Lakes with explorations of what I suspected was an old mining road. It was clear it was from logging. It was also clear that although a few trees have come down, it’s being kept open for hiking. It certainly isn’t usable (or legal) for driving. I decided to add the system as a path. It’s outside the wilderness, so the bikes can use it too. Apparently I was almost to the end when I turned back.
I was surprised to see that the ATV trail hasn’t been mapped. I marked the bridge and got it a little further, but then it gets too close to the creek and the creek is often not in the right place and I got frustrated. It goes through to somewhere and connects to another trail that climbs soon after where I stopped. (That trail is also missing.)
Logopädie List Lister Meile 85 a 30161 Hannover
Logopädie Kleefeld Scheidestraße 21a 30625 Hannover
Logopädie Hainholz Bömelburgstraße 35 30165 Hannover
Logopädie Bothfeld Sutelstraße 10 30659 Hannover
Logopädie Burgdorf Weserstr. 6 31303 Burgdorf
Langenhagen Walsroderstr. 47a 30851 Langenhagen
So I attacked the dreaded West Elk. I think I started faltering on marking trail visibility near the end, but I started off well on Coal Mesa. I marked the camp good camps. I didn’t mark the spring I found to camp that first night… Maybe I have to go back. I made sure the trail was really clear around the peak, which has some problems. Stay low, whatever you do! It doesn’t look like much, but it goes very directly for the last 40 feet.
I didn’t add any of the trails I didn’t see anything at all of, and there’s a bunch. I did make sure everything off the side of North Baldy was marked. I managed to connect it to the trail even. Put down some cairns. It looks like it probably connects to Beaver Creek far down rather than going around the top of the bowl that Beaver Creek occupies.
All that informal stuff around West Elk Peak is now marked as such and has difficulty and visibility. There’s some trail visible down low on the evil T4 track going north from the peak, so I decided not to give it visibility=no.
Added some more camps I’d noted along the way. It’s good info, it is. I couldn’t note the no camping. There’s quite a few lakes that have no camping allowed within a quarter mile, so it is something that is needed. Google was uniquely unhelpful deciding I was on about subjects that have nothing to do with camping.
I marked some of the trail I found as I left Sheep Lake (and the nice camp along it!). I adjusted the junctions into something sensible that at least resembles what’s on the ground. I marked the south route down as visibility=no and added a note that I didn’t see any evidence it was ever there. I did find a track on the other side of the lake that wound around to the trail which included one blaze and a cut log. Nothing at all for the larger trail.
Before going to Dillon Pinnacles, I noticed one map shows the trail to the pinnacles (and a couple interpretive signs) while another shows a trail up Dillon Gulch. This trail predates the reservoir and just goes to a spring. (The map actually showed it just randomly starting at the water with no entry other than boat.) I found it signed and it is easy to trace all the way to the spring and no further as advertised. The sign was specifically giving the dates. The land between the NPS and USDA is state wildlife, I believe, and they don’t want you up there for the big game migration in late winter, early spring. And now you know from the map.
I hiked Ptarmigan Lake and pushed on to the nearby peak. The last part is cross country, but the since I did it by the road, I did almost all the trail. Since the road crosses the trail, I decided it would be good if it was mapped properly. The trail was marked T4. There is absolutely no point at which you need to get your hands out of your pockets for this trail. T3 is really pushing it. Everyone travels the trail next to the lake, which has huge steps and leaves the lake on a very steep slope. I noticed the higher trail (which one person was taking, so not quite everyone) that turns out to be very smooth and obviously the built one. I marked it, but only as old trail. Correctly, it should be marked as the actual trail.
There’s rumors there’s work happening on Lost Lake, so I didn’t touch that.
The picnic area over Cottonwood Pass has a well used trail to the edge of the wilderness. The old picnic area had a much larger trail that I found as I hiked about. There was a break and then obvious trail again as I hiked out to a big shoulder. I decided against adding this. My track is public for folks to see, though.
This hike up Turner Peak started along an informal trail that someone put up roughly. Might as well get it better. I didn’t bother with the rest of the track, but some of that on to the peak was very trail-like. Not so much my travel over to South Texas where the CDT travels now. I left it alone.
Harry Woods posted a diary entry a couple days ago about his attempt at editing every day for a year, and I started wondering what the records are for the most number of consecutive days that people have contributed to the map. I already had some code lying around for parsing the changeset history file so it was pretty easy to mine that data. It turns out there are some super dedicated mappers! Here are all the people who have editing for at least 1000 consecutive days:
Edit: Fix an error in table calculation (Thanks Aurimas Fišeras!)
In einer Korrespondenz auf dem issue-tracker eines Routers auf Basis von openstreetmap Daten bin ich auf die default Einschränkungen für die verschiedenen Arten von Wegen hingewiesen worden.
Im lokalen Forum wollte ich daraufhin anregen, dass die Berechtigungen von track auf etwas der Realität näheres als ein vollmundiges yes gestellt werden.
Ein Vorschlag eines Teilnehmers aus einem Drittland war partial, was meiner Auffassung so ungefähr das bedeutet, dass es vor Ort noch erhoben werden muss. Dergleichen finden sich aber nur auf zehn Prozent der als track erfassten Wege. Trotzdem hätte ich beinahe geglaubt, dass das auf ganz Österreich gut zutreffe.
Offensichtlich ist das nicht so. Von mancher Seite wird der Standpunkt vertreten, dass wenn vor Ort nichts ausgeschildert ist, dann ist yes sehr wohl richtig. Mir entzieht sich bis dato, was das damit zu tun hat, ob in den openstreetmap Daten eine Einschränkung erfasst ist oder nicht.
Nun ja, es könnte in der Tat in den meisten Fällen so sein, dass wenn nichts erfasst ist, auch nichts ausgeschildert ist. Das allerdings deckt sich nicht mit meinen Beobachtungen in Tirol.
Wie das nun belegen? Sind ja hier wie dort nur Anekdoten! Also amtliche Daten geholt, den Intermodales Verkehrsreferenzsystem Straßengraph, a.k.a die GIP für mein Bundesland.
Das Shapefile in JOSM laden dauerte ewig. Wenn einmal geladen lief das Schieben des Ausschnitts aber recht flott. Super Sache dieser Editor. So lässt sich gut beurteilen, wie die openstreetmap mappings und die Kategorien in den amtlichen Daten korrespondieren. Sie tun es, mit Unschärfen an den Rändern zwar, aber Ja!
Dann entdeckte ich das geojson Format und kam auf die Idee, damit wäre es möglich, die Daten direkt auszuwerten. Hier zwei jq Einzeiler:
Un metrominuto es un mapa esquemático de un municipio o urbe que representa las distancias entre los principales puntos de la misma y los tiempos medios que se tarda en desplazarse caminando entre ellos. Su diseño a base de puntos y líneas de colores los hace semejantes a los mapas de transporte público habituales en las ciudades.
Esta entrada describe los pasos a seguir para elaborar de forma sencilla un metrominuto de tu ciudad o zona de interés.
Selecciona un grupo de puntos de interés en la zona objeto de tu mapa, según su importancia o grado de interés para el público objetivo. La distancia entre puntos debe ser apreciable, si bien se recomienda que no sea mayor de 4 ó 5 kilómetros.
Con este primer paso, ya podrás conocer si tu mapa tendrá una disposición vertical u horizontal. Puedes usar herramientas en línea para la selección de los puntos sobre un mapa real de la zona (p. ej., Google Maps o uMap).
La conflation, également appelée appariement ou fusion de cartes, permet d’obtenir un jeu de données à partir de plusieurs. Cela vise en autres à une meilleure qualité, et à limiter la redondance des données. Différents outils et méthodes permettent cela sous OpenStreetMap, dont une extension JOSM.
L’idée générale est simple, il suffit de prendre les deux jeux de données et comparer un à un chacun des éléments de part leur géométrie (position spatiale et forme) et caractéristiques. Manuellement, comparer des données spatiales revient à avoir chaque jeu de données sur un calque et superposer les deux pour mettre en évidence les différences géométriques avant de comparer les caractéristiques de chacun d’eux et voir si une version est plus précise que l’autre, ou si des données concurrentes apparaissent. Numériquement, le même principe est utilisé, seulement il est grandement facilité via des scripts automatisés qui apparient les données selon différent critères. Mais bien sûr, rien n’empêche de le faire manuellement avec les données sur deux calques, ce sera juste très long et répétitif.
L’extension repose sur la fonction « Remplacer la géométrie » de l’extension « UtilsPlugin2 » qu’il faut également installer. Cette fonction permet de fusionner les caractéristiques de deux éléments en ne conservant qu’une seule des deux géométries.
Les deux jeux de données auront un rôle de « Référence » et de « Sujet ». Le premier correspondant usuellement aux données à importer et le second aux données OpenStreetMap.
Par simplicité, je préfère travailler les données en dehors de JOSM via un tableur ou des commandes de remplacement. Il s’agit alors de transformer les données de manière à coller aux attributs OpenStreetMap en perdant le moins possible de détails. Les puristes le feront sous format GeoJson, Shapefile ou que sais-je encore, je suis un homme simple, j’aime les csv.
Added description to car park surface and its entrance.
Added the existence of Solar Panels using ‘Every Door’ software on iPad.
I had previously tagged data about the tea room in the parking area instead of the building. Today I have moved the data to associate with the building which I have also re positioned and added a brief structural description to.
I have also added and area which is the car park though the outdated Bing imagery is confusing the layout. At least it has a parking area marked. I have to go back and mark the existence of the new solar panels on the south facing roof pitch which have been installed over the last week. I have no details on capacity, batteries etc. at this point.
As expected, there’s not much to worry about on the Mineral Belt Trail. It had some signs missing and has a few less now. There’s a picnic area that wasn’t marked. It was a boy scout project. (Says on the sign.) It includes sighting tubes for the local mountains! Ah, but how do you mark those? I gave it a go.
Ah, the Ute Peak saga. My first hike in the area was almost, but not quite, up to the top of Ute Peak. Then I wandered up Darling Creek. Then I came at Ute Peak once more, but from a long way around. The trails connecting the two South Fork Trailhead were something I really wanted to know about, but FS and OSM were keeping quiet. The only thing in the area mapped on OSM was a strange alternate CDT route that vanished whenever I tried to zoom in. I hiked on the Forest Service quads. They had some strange ideas for Ute Peak, but not half as strange as I found when I got signal up high and got the USGS maps of the area.
Anyway, I started mapping by adding the Darling Creek trail up to Saint Louis Divide. This is one of the South Fork Trail’s trailheads, so I figured get it done first so South Fork can build on it. I also tried to get Saint Louis Divide on. There’s not a lot of trail to the trailhead on the east and it can be seen on satellite. There’s some really good game trails in the area, too. They can be the easiest trail of all to see in this area. I’m not mapping them, but did seem to get a little obsessed with mapping this area that is so undermapped. (Denver, what you up to? This is your backyard!)
Then I added Ute Peak. I did it via the Ute Peak Trail, but the more common route is from Ute Pass. USGS has a pair of lines that eventually get to Ute Peak somewhere entirely different from the actual junction. I used the Strava heat map background on Strava-iD to find the trail.
Apparently someone was in the process of getting all kinds of trails down on Medicine Bow around the peak when my Wyoming map file was created. (OpenAndroMaps updates on a roughly quarterly basis.) There were a lot on my map and a bunch more once I started looking at what could be mapped from my excursions. I added some signs. The area is well signed. And I noticed that the main trail up the peak (the east side of Medicine Bow Peak Trail) was of the right shape, but the wrong location. That seemed odd, but the Strava heat map confirmed my line was in the correct place. There was a line one could imagine was trail under both lines. So I moved it. Maybe it was a USGS line? Maybe I should more carefully check the rest? It seemed alright on the first look.
There’s some amenities that could be added. A hand pump for water in a picnic area, for instance. But mostly things are looking good for the area.
I decided against trying to mark the old Circle Trail that I foolishly tried to hike. I’ve seen people mark all the stops on interpretive trails like Miners Cabin, but I decided against trying that.
In this area, the Encampment River isn’t even on the map! I didn’t do much to fix that… I did add the southern half of the Encampment River Trail, which happens to be the part I hiked. It’s more popular than not being on the map indicates and now it’s complete on the map.
Before Encampment, I hiked in the north portion of Mount Zirkel Wilderness. There were no trails here on OSM. There were a couple track roads. In the congressionally designated wilderness. Not on my watch! I actually hiked quite a lot of the trails here, and added a few more based on what can be seen and USGS. It seems to be fairly accurate in the area. I was able to adjust a lot of these to trail visible in satellite pictures.
I do have a difficulty here for the trail visibility. If you read my blog, you’ll find a theme. After the first 5.5 miles, I launched into two miles of the most obvious but difficult trail I’ve ever encountered. In the middle of the second day, I launched into another 2 mile section that was much the same. On the third day, I headed out into another 2 miles that was definitely going to be worse than the lower trail, defiant about if it would be. There was a bit that was the worst piece of trail for the whole trip, but it was a bit shorter.
