There's a place for being self-centered in art-making. Ultimately, I make most of my personal art for myself, even if the stated purpose is to use it as a gaming aid. My players honestly don't mind if I'm sketching a map on a napkin during the game, so when I'm going to the trouble of creating something fantastic, it's only partly for their benefit.
We do talk a lot about verisimilitude around here because having a logical environment with realistic underpinnings helps facilitate the willing suspension of disbelief that fantasy gaming requires. It also makes things a little easier on the GM because when your fantasy world works like the real world, you don't have to stop and think as much about how an unexpected player action will affect things.
Now, given that we do spend a lot of time discussing how our maps should reflect real-world physics, ecology, and climatology, every once in a while there's a bit of push-back where people say, "Can't we just have a really fantastical fantasy world? Is that too much to ask?" A general thread like this one is an ideal place to have that conversation since it's usually not appropriate in a Work in Progress thread.
Being able to use my understanding of how the "real world" operates is nice. It's useful to be able to apply the notion that fluids flow downhill, even if that sort of behavior would fill up the local Underdark in next to no time. I like the idea of knowing that fires will consume things, even if I can't actually cause 40 foot diameter balls of fire to appear and set everything only in that area on fire while leaving the things next to it unscathed.
Suspension of disbelief is critical for most entertainment activities. Requiring that everything and everyone operate according to exactly one standard of efficiency kills the sense of wonder needed for enjoying a lot of the world. Silly-looking and impractical buildings are central to lots of stories. Massive delvings that would take a thousand years to dig are also central to lots of stories.
Peculiar architecture in the real world helps people to remember a place. The opera house in Sydney is a seriously silly-looking building. So's the Eiffel tower in Paris, the TransAmerica building in San Francisco, the rock-cut churches in Lalibela, Ethiopia, and even the pyramids in Egypt. But each of those places is fairly well-known because of their iconic architecture, not because of the uninteresting and more efficient buildings nearby.
Modern architects are desperate for their works to be iconic, and are getting increasingly desperate in their works. Is this trend a good thing? It depends on what you're trying to optimize for and how much you're willing to pay for it.
Try not to get the notions of real and consistent mixed up. If teleportation is cheaper than maintaining a road network, there won't be much of a road network on the map. If super-strong materials are cheap and readily available, then architecture may look nothing like something evolved from a mud-brick wall. Basic physics will probably still work, but with minor exceptions (water flows downhill except in certain places, roads will follow low-energy paths because people are lazy, places for boats to dock will be on bodies of water, and so on). It's the basic framework that we bring with us from the real world that allows us to read the map in the first place, and it's up to the fantastical mapmaker to decide how (any why) to have elements on the map that don't jibe with preexisting knowledge. If that makes any sense.
Just a thought I asked my architecture friend about this and he suggested looking into Frank Lloyd Wright's designs. I remember visiting one of houses designed by him in Ann Arbor MI (no straight corners) simply enchanting.
Having just visited Toronto and visited two museums with newfangled additions to them I would say you are starting your argument from the wrong place. Aesthetics are not exclusionary to form nor is the reverse true either. It does take a person with sensitivity to both to marry them together as a working whole however.
Based on my experience Daniel Libeskind is not that person however. The new addition to the Royal Ontario Museum is an uninteresting blight on the outside of a pleasing but fairly standard building, but it's inner spaces are just downright terrible and awful. They force poor use of space have a bunch of closed off pointless views and ledges and the pathing to get around inside is poorly thought out. And his "specially" designed chairs were like sitting on the top of a slide while trying not to go down. For him I suspect his architecture and his feeble attempts at design really are all about ego. I not that he also did the Denver Museum mentioned earlier in the thread apparently.
Just down a little ways from that how ever is the Art Gallery of Ontario (Frank Gehry). This expansion added a relatively simple glass box extending the building up another two stories and a new glass 'sail' facade. From the outside it looks fairly new and interesting and it doesn't obscure any previous older buildings from view (largely since earlier additions already took care of that). More to the point the inside was really, really well done. It made lots of sense in human terms and worked with the natural flow of expected use. It has a feature stair case which normally I'd expected to hate, but really I found my self drawn to go up it just for the experience. None of the views or interesting things were cut off but rather expanded upon and framed to draw you onwards. Sadly the 4th and 5th floor were closed to set up an exhibit so I didn't get to visit them but given how well the rest of the inside was renovated and how well the spaces flowed together I would expect more of the same.
Behind the AGO is the New OCAD building which is just a square block set up on stilts. It's not especially pleasing to look at on the outside but it is at least interesting, and inside it is what you would expect to find in a similar box located on the ground if it was designed for a school. Nothing spectacular in any way other than the fact it starts 9 stories up. That said it's insides are not conduits of functionality, flow, or efficiency, it gets the job done but doesn't improve on it in anyway from any other building.
The point being that regardless of the aesthetics of a building its functionality cannot be determined from it's outer appearance. Nor is a fantastical or strange outside indicative of an inefficient building or functionality. Indeed some especially efficient buildings I've seen pictures of look very strange indeed. So the thing to do is make sure that the map or dungeon or whatever makes sense to you within its aesthetic frame work. Also remember people (or dwarfs, or elves) build and makes things for a plethora of reasons and desires of which functionality is merely one (and once taken care of perhaps one of the least in many cases).
Originally Posted by Altrunchen
So how does one overcome the urge to make map that makes sense and feels believable when attempting to make a fantastical map?
Fantastical could mean off-kilter with respect to our norms, or unusual with respect to any conceivable frame of reference :-). To be pleasantly fantastic one could devise a self-consistent system, and have your world stick true to that. Because by fantastical map, you really mean fantastical *world*... the map not being fantastical on its own unless you are drawing with rhubarb on lettuce leaves, or building a five-dimensional globe from frozen smoke and diamond chips.
Being consistent within a defined framework is good solid worldbuilding, that your players or readers won't have to constantly figure out.