I know the issue here has already been agreed, it's a bug.
I'm veering OT, but for the mathematically minded, I still have a nitpick or perhaps interesting point, considering the subject of touching. The thing is: I believe at most two objects can touch at any given point. This is true in all (numbers of) dimensions. Why, you might ask?
Well, the thing is, there is no infinitesimally small physical object, such a thing does not exist. There is a smallest unit (a geometrical body) that can exist,
an atom, if you will. A "point" does not make sense when dealing with physical bodies.
You can think of this in many ways; say, if two corners are meeting each other, can air pass between them? If yes, they are not touching (since air can pass). If air can not pass, they must be touching. But remember the air is also "a physical body", if air can pass, the two bodies of air (as divided by the rooms) are touching! Both can not be true at the same time. See:
Another way to think about it is, say we have a surface (call it A). Cover it partially with a wedge-like body B (it doesn't necessarily have to be wedge-like, but makes easier to get the point trough). Take another similar body C, and cover rest of A with it, so that it's touching B. Now all A, B and C are touching each other. Take a body D, and notice you can not touch A with it! You can take B and C apart a bit and now A can touch D. Bot not both sets A,D and B,C can touch each other at the same time (at the same place).
You could also think of a pie and slice it into arbitrary many pieces so that the cutting line intersects at the same place. The pieces are only touching their neighboring pieces (at most, or exactly two pieces, if the pie is whole). If the cutting lines are not perfect, then two (and only two) pieces can meet at (or around) the center point.
This is not just a physical thing but also a mathematical thing; it depends on how we define touching. If touching means the areas need to have more than an infinitesimally small point of contact, then at most two objects can touch at a single point (or around; more precisely on an [N-1]-dimensional plane at this point in N-dimensional space). I don't know enough about the subject but there must be some theorem, axiom or something in the field of topology or maybe even set theory, linear algebra or something out there.
EDIT: By this analogy, no pieces taking up spaces on a grid are touching only by their corners, as the point where the corners meet, is infinitesimally small.
). But the explicit mention to corners makes it much clearer (although it says that corners can touch, maybe corners should not be considered "part of a wall"?).
