[Tagging] Rooftop parking -> new parking=rooftop value?

johnw johnw at mac.com
Wed Nov 12 01:16:16 UTC 2014

> First Principle? 

> However in a multistory buliding .. what are people coming to the building for? Should 'we' not map the purpose of the building

The purpose of the building is indeed retail (almost always), but the purpose of the map is navigation. 

I wish to accurately tag and render something you would see on your GPS while driving in the vicinity of the mall. Some of them are quite complex - I’d like to think I have a very good idea of my surroundings, but while trying to leave a very large and complicated shopping mall rooftop, it was impossible for me to visualize the warren of paths on top of the roof of the building I was on. I I was trying to exit on one side, ended up exiting on another. 

> Provision of parking may be important .. but it is not the most important thing. Only a single purpose parking complex has the sole function of providing parking.   

Yep which is why I thought having a tag parking=roofttop for an area tagged as amenity=parking that is (almost always) contained *inside* an area tagged with building=* would warrant different tagging or rendering. 

> Second .. Effort? 
> A miltistory thing will have many surfaces, one floor may be concrete, another carpet, another stone. Tagging these in some sensible way may be fine

All for it. 

> .. but how will they be used? 

shop=mall is a good place to start. Indoor mapping is really difficult because of the scale of building vs the road system we are mapping, and the overlapping nature of multistory buildings, so a system hasn’t been developed into something useable for a basic mapper yet (IE; me, mapping in iD), and considering the difficulty of turning a building outline into an accurate representation of it’s internals, it will almost impossible for for satellite imagery tracers. 

BUT - This is one of the few cases where the road system actually goes on top of buildings. There are few other cases where a common class of buildings that  are *under* a road system (there are thousands of surface rooftop parking lots in Japan, and many rooftop multi-storey ones) and still have the building itself be visible for a navigation/mapping purpose. 

These roof parking lots seem like an easy thing to tag, and is still unrelated to what is going on inside the building. 

I thought changing the opacity of it's render would give more priority to the points with shop tags / indoor tags that will eventually populate the building, but would let the parking layout visible on the roof for car navigation. 

> And is it worth the effort? 

It’s about 10m for me to map out a single parking lot, to go with the hour or so I would spend mapping the basics of a small mall. I tend to spend a lot of time mapping retail places, because I assume the detail will be very useful to people wanting to use a map to navigate to and around large retail establishments, such as the rare large malls here in suburban/rural Japan ( ~ 2 million people with 3 malls in my prefecture, and ~ 15 large shopping centers.) 

> Considering the quality of the map in some parts of the world would not these efforts be better spent in improving the map elsewhere? :) 

I’m turing the soup of unaligned & outdated unclassified/residential roads into something more usable in my neck of the woods - getting some more tags isn’t going to slow me down too much - I just want to describe what cars and pedestrians are going to use, or places they want to visit. ^_^

Now that the tagging structure can handle roads, driveways, tracks and trails with very high levels of detail, the effort to refine the tags for even smaller and more local conditions, along with more detailed tagging of areas is becoming the purview of the tagging groups issues, right? Until there is some drastic change to the main structure of the tagging system, the future of tag refinement will be ever more increasing tags for detail of in missing areas, and refining and codifying the messes of existing tags to make tagging/parsing easier.  Right?

The frist time you run into a tagging situation that seems unwieldly or missing, and decend into the wiki to figure out how to tag something, it really knocks the wind out of new taggers. Train stations nearly sent me packing.

If we have enough detailed tags to handle *anything* taggers can find - codified and explained on the wiki, properly made presets for them in iD, and renders in  -carto, the world will eventually be mapped to our satisfaction by an increasing number of amateurs.


More information about the Tagging mailing list