I have an ongoing debate with a friend and I want to get some additional input on this question... For a shopping center or mall, my friend insists on mapping the shops inside by guessing the outline of the shops, drawing them as individual areas, and then piecing them all together to form the outline of the larger building. I've been insisting, on the other hand, that one should never make up data like this (or even make educated guesses with such frequency), and that, instead, the best thing to do is to draw the known outline of the building and then place points within the building area in the approximate locations where the shops are. My friend has countered my approach by saying that this violates the "one feature, one element" rule, since a commercial building area with several points inside would be several elements for a single feature. However, I believe this is a misinterpretation of the rule. As a sort of compromise solution, my friend proposed making each shop a node on the perimeter of the area of the building, thus associating the shop information with the larger building building. What do you think?

I have an ongoing debate with a friend and I want to get some additional input on this question... For a shopping center or mall, my friend insists on mapping the shops inside by guessing the outline of the shops, drawing them as individual areas, and then piecing them all together to form the outline of the larger building. I've been insisting, on the other hand, that one should never make up data like this (or even make educated guesses with such frequency), and that, instead, the best thing to do is to draw the known outline of the building and then place points within the building area in the approximate locations where the shops are. My friend has countered my approach by saying that this violates the "one feature, one element" rule, since a commercial building area with several points inside would be several elements for a single feature. I believe this is a misinterpretation of the rule. As a sort of compromise solution, my friend proposed making each shop a node on the perimeter of the area of the building, thus associating the shop information with the larger building building. What do you think?

I have an ongoing debate with a friend and I want to get some additional input on this question... For a shopping center or mall, my friend insists on mapping the shops inside by guessing the outline of the shops, drawing them as individual areas, and then piecing them all together to form the outline of the larger building. I've been insisting, on the other hand, that one should never make up data like this (or even make educated guesses with such frequency), and that, instead, the best thing to do is to draw the known outline of the building and then place points within the building area in the approximate locations where the shops are. My friend has countered my approach by saying that this violates the "one feature, one element" rule, since a commercial building area with several points inside would be several elements for a single feature. I believe this is a misinterpretation of the rule. As a sort of compromise solution, my friend proposed making each shop a node on the perimeter of the area of the building, thus associating the shop information with the larger building building. What do you think?