hladno_pivo wrote:sarebt wrote:... I'm not a plotter, so can't help you directly. ...
What about indirectly?
I do what I can
hladno_pivo wrote:sarebt wrote:hladno_pivo wrote:
Some famous Bohinj mountain or alpine pastures (planine) should be merged with the correct profiles of the villages in the valley bellow. However, the GPS is not the same for villages and the pastures. [...]
...Respective locations are then defined as combinations of settlement+postalcode. But in Slovenia this leads in most cases to 1 coordinate as settlements generally have 1 postalcode only.
None of the three seems to be an official settlement (naselje), so they should be merged into the profiles of the cities on which territories they are without coordinate assigned. (that is: they will get the average coordinate).
OK, this is not true! I will list 3 reasons / facts why this can & should be done.
1. Precedential cases. There are many similar cases with one city profile (and postal code) but with different GPS tags. Please see "Vogel" (ski resort and not an official settlement). You will observe that it has different GPS coordinates than the village "Ukanc" (with only one city profile for both,
i.e.,"Ukanc"). There are more cases (among others, Šmarna gora in Slovenia is similar case).
2. My suggestion is closer to reality and more correct than the one suggested. You suggested average GPS coordinates!? Not wise. Average coordinates would put the locations in the middle of nowhere.
3. For the sake of dot hunting.
Best solution: different GPS, one city profile.
1. I'm not sure if there are any precedential cases. Only based on the existence of the profile, I wouldn't be able to say it's a precendential case, but just a similar one (I can't see the difference in coordinates I think as I don't have notes from it, but immediately believe you).
2. When carried out consequently, the coordinates will not land in the middle of nowhere, but in the village that the note is part of. There is only 1 coordinate per village as there is only 1 postalcode in a village. The infrastructure on Vogel is officially nothing more then a street of Ukanc, although it maybe feels different. But see my question below.
3. This is surely not an argument, dots are based on coordinates, not the other way round. You cannot define locations purely on the fact that you need a dot somewhere

, there are on average more than 150 settlements per dot in Slovenia, I'd say chances enough.
Nerzhul wrote:The new data structures are organized this way:
We have cities (parents) and locations (children). Locations are defined by a cityname, postal code and country, e.g. 'Bamberg, 96047 Germany'.
http://forum.eurobilltracker.eu/viewtop ... 690#347690
What would be your objective criterium for defining the "cityname", if not by the officially defined settlements?
hladno_pivo wrote:sarebt wrote:hladno_pivo wrote:Most complex error, there are several cities summarised in the same profile:
1. "Bohinjsko jezero". It is the name of the Bohinj lake where three settelments are found: Ribčev laz, Stara Fužina and Ukanc. To add to confusion also the post office is named "Bohinjsko jezero". There is no such city / village with this name. Post office is in the city Ribčev laz so I suggest:
Bohinjsko jezero --> Ribčev laz (postal 4265).
2. As far as I am aware, if somebody wrote "Bohinj" , which is a municipality with close to 20 cities / villages, it was automatically assumed that this was an unexisting city "Bohinjsko jezero". I recommend a solution, which is to assign the postals to one of the three cities in the area 4264 (Bohinjska Bistrica), 4265 (Ribčev laz) and 4267 (Srednja vas v Bohinju).
Yes, those are difficult cases. I would say the location of the post office doesn't really matter.
It sure is a contributing factor and it should not be ignored so easily without argument. This is the only post office (and also a bank) in the vicinity of the lake (btw: all hotels, sport & tourist lake activities, dozens of bars and a disco club are in Ribčev Laz). Stara Fužina is the second biggest village with only one bar near the lake. Basically it is hard to spend anything out of Ribčev Laz and still by the lake.
Ok, I thought you were talking about the postoffice because of involvement of the postalcodes. But economic center, fair enough.
hladno_pivo wrote:sarebt wrote: My suggestion would be to unassign these notes at all as they refer to a non-exisiting town, but merely to a region and you cannot be reasonably sure which of the places they were received. However I don't know what the official procedure is in this.
Let's not replace a good mistake with some other and even less accurate construct, please.
Unassigning the notes is not a good solution. They should stay as they are. Only the city profile
"Bohinjsko jezero" should be merged with "Ribčev Laz". That's all.
After looking to this again, yes maybe I was a bit too rigorous here. At least for Bohinj 4264 and Bohinj 4267, assignment to B.B. and Srednja vas seems to be logical. And the others, well yes, maybe in this case it's a better solution then we have now, even if not perfect.