OpenStreetMap
giggls's Comments
| Changeset | When | Comment |
|---|---|---|
| 149488066 | Es stimmt, bei tourism=caravan_site ist das nicht ganz so uneindeutig wie auf Campingplätzen aber es trägt trotzdem nicht zur Klarheit bei. Spätestens dann wenn man zwischen capacity:caravans und capacity:motorhome differenzieren möchte ist das auch hier nicht mehr eindeutig.
Übrigens gibt es beim mappen von tourism=caravan_site und tourism=camp_site noch jede Menge *erheblich* sinnvollere Beiträge zu leiste. Siehe mein Blogpost vor ein paar Tagen:
|
|
| 149404477 | Der andere Platz ist ebenfalls diskutabel. Sobald es vollwertige sanitäre Anlagen mit Dusche gibt tagge ich eigentlich nie backcountry. Selbst wenn der Ort jwd ist. |
|
| 149404477 | Moin, als Author von https://opencampingmap.org kenne ich mich ein wenig mit tagging von campsites aus :) camp_site=basic ist übriges etwas was ich komplett ignoriere, weil das IMO ein hoch subjektives Kriterium ist. Der Platz erfüllt IMO jedenfalls nahezu keines der Dinge die ich für backcountry voraussetzen würde. Im Gegenteil: Es gibt vollwertige sanitäre Anlagen und die Lage ist nicht weit ab sondern city-nah. Was ich stattdessen taggen würde:
|
|
| 149404477 | Why did you revert my change? This is *not* a backcountry site. |
|
| 147687566 | ||
| 147687566 | oK, eventuell noch ein description Tag dazu machen, dass der Platz derzeit geschlossen ist. |
|
| 147687566 | Hallo,
|
|
| 141987508 | Well basically cycle-routing *did* lead us this way:
BTW only reason I went looking at the data exactly at this place afterwards because I think this was not a good idea. In practice it was impossible for us to cross here reasonably fast with our 20kg of luggage when travelling by bike trough the baltic states by bike last year. I will thus see if I can modify bikerouter profiles to ignore routes passing crossing=no nodes. This said it looked completely crazy for me from my German point of view to map such a thing as any kind of "crossing". |
|
| 141987508 | Well access tag is about what is allowed to do not about what is possible to do. I do not know about the legal status of crossing railways in Latvia but it is completely illegal to cross a railway at spots without a corresponding traffic-signs in my country (Germany). A very similar example would be tracks and paths ending at a motorway where it is strictly illegal to use them for the general public. An example would be this way:
Regards Sven |
|
| 141987508 | Of cource this is Latvia but nevertheless this did not look like on the ground like an existing path either.
|
|
| 141987508 | While it might be common to map illegal crossings in Lithuania it is definitely not in my country.
Regards Sven |
|
| 124561287 | Sorry, I do not speak french very well. The sites might well get re-opened this year.
|
|
| 126035627 | I do nor *remove* the site relations. I do evaluate them in a meaningfull way which means that they do not represent two camp-istes but onlytwo aspects of the same site. |
|
| 126035627 | Oh now I seem to understand what you did not get yet. I do not consider a site relation representing an actual geometry of a campsite itself but forming a couple of scattered objects related to *the* site which is a member of the relation. Thus in my map I just assign all the related objects types as a feature to the camp_site inside the relation. My database table for drawing campsites does not contain site-relations anymore. They are just used for adding actual features to the sites (area or node) themselves. E.g.: Site relation contains a restaurant -> Add restaurant=yes to node or area member tagged as tourism=camp_site This is what breaks my map. |
|
| 126035627 | Will I really need to make a list of what is broken in all those objects. However, it is quite easy. All the objects outside the polygon or in case of campsite_points all of them are not shown in my map as features of the campsite anymore as they are no longer related. E.g. Hollenbacher See has no fast-food and no playground. Oh and they are not part of the fenced campsite area thus mapping them as inside the polygon is simply wrong on the ground. Frankly I do not think that it is a good idea to use the "One feature, one OSM element" practise as a dogma if violating it does make sense as it is the case here. P.S.: I would like to end this discussion in the changese and try to clarify this issue on the tagging Mailinglist. See my post there:
|
|
| 126035627 | Did you even try to understand the problem I try to solve using site relations? Did you read my blogpost? You broke my map and you are not providing me a reasonable tagging Alternative :( Camp-sites with external features like showers, toilets etc. do exist in the wild and I think it is good to have a solution at hand to map this. |
|
| 126035627 | We are not talking backcountry only sites here but tourist sites also. I will show you another example from the ones you broke in my map. Have a look at the following site relation (currently with tourism=camp_site tag removed by your changeset):
The polygon here is clearly a camp-site rather than a pitch. In fact josm would even give me errors if I would change this to tourism=camp_pitch because there would be lots of individual pitches inside a giant pitch then. The members of the site relation are part of the camp-site as well but are located outside the polygon because they are also open to the general public. I do not insist in taging the site relation *and* the polygon as tourism=camp_site but I do not know a better solution. Abusing camp_pitch is not such a solution IMO, given the fact that the reason is compliance to a wiki page which even states that the principle they talk about there "is not absolute". Fact is that my map is currently incomplete[1] because you removed the camp_site tagging of the site relation.
I am open to better tagging ideas, but tagging camp_pitches inside camp_pitches is definitely not such a solution. [1]Fast-food and playground are currently missing from my map. |
|
| 126035627 | BTW I don't think that we are this far away in our thinking.
I did even make a bug page showing evil camp_site in camp_site tagging here:
BTW, I do show site-relations which contain more than one camp-site object as a bug on https://opencampingmap.org but having _one_ object tagged tourism=camp_site inside a site relation also tagged tourism=camp_site is definitely not something I would consider a bug. |
|
| 126035627 | Sorry to say, but this is crap.
Would you please read my blogpost which describes the rationale behind the site-relation support in OpenCampingMap? I did not revert your changeset I might have fixed a few of the sites where I realised the missing tourism_camp_site tag which I considered a mistake on my site up till today when I discovered this changeset. The wiki entry you are referring to is IMO not applicable to site relations. Unfortunately the site-relations are useless without the tourism=camp_site tag now as I can not distinguish between camp-site relations and other site relations without them anymore. An Object may well be a member of more than one site-relation with different type which is perfectly fine. osm.wiki/Relation:site clearly states that site-relations "must have a main tag defining whatever feature the site relation describes" You did actually kill a feature of my map by this changeset which I am now unable to support. |
|
| 126035627 | Jetzt erst habe ich diese Änderung hier gesehen. Ich hatte schon an meinem Verstand gezweifelt. Dieser Changeset hat die Darstellung aller geänderten site-relations in der OpenCampingMap zerstört weil Du die Zuordnung über tourism=camp_site gelöscht hast :( Würdest Du das bitte rückgängig machen? Hintergrund ist hier in meinem Blogpost:
Die dort gezeigte Relation hast Du ebenfalls kaputt gemacht :( |