That's without even going into whatever makes your world "fantasy". Un-Earthly geology could be plot relevant (The Fifth Season, N. K. Jemisin). Or it could be a bit of colorful background that's still scientifically valid. Maybe you live on a tidally locked moon, or a geologically dead world with a cold core, or an ocean world with an ice crust (not of water necessarily). Maybe the locally important geological phenomena come not from inside the planet (plates, volcanoes) but outside (asteroids, meteors). Maybe your planet's liquid core has currents so strong that islands float around at meters per second and bounce off each other.
Fantasy only needs to be consistent with yourself - not with our Earth. Otherwise where's the fun in worldbuilding?
Relevant article from the same site: https://mythcreants.com/blog/should-your-fantasy-world-resem...
I think it's only fun if you operate within constraints. Choosing Earth-geology is a pretty reasonable constraint that allows you to more quickly get on to other aspects of worldbuilding (the ones that most fantasy worldbuilding focuses on anyways). What you are describing sounds more like science fiction worldbuilding! Plenty of room for overlap and deviation, of course.
Authors often build geologically inconsistent worlds. The article has suggestions for fixing that by changing the geology to match Earth, but that will force the climate etc. to match Earth too. I think if an author wants to build a world with a geology different from Earth that's not a bad thing - they should just notice this and make sure it's consistent and has an explanation for being the way it is.
A formula romance cannot solve the problem of a long standing antagonist by having the hero grab a magic wand from nowhere and kill him - readers of romance generally assume a very earth like world because the story is about the romance and changing the world to add and explain magic is a distraction. So the standard course of action is for the hero to use a gun - guns already exist and so very little explanation needs to be spent on where the hero gets the gun before the story goes back to the romantic parts the reader is interested in. That isn't to say the hero cannot use a magic wand in a romance story, but magic needs to be introduced by page 2 at the latest - and some readers will decide that they don't want to read your book because magic is involved.
Note that non-fiction doesn't have that constraint. I can introduce the magic wand used to kill right at the moment in history where the magic wand was used. Readers will go "what", and do some double checking and discover that a magic wand was actually used (more likely the research will discover that nobody knows what was really used, but it looked to observers like a magic wand so that is what everyone says), then you will go back to my story satisfied that it really happened that way despite being unbelievable.
(There are probably exceptions. Douglas Adams could make a lengthy digression on geology interesting, amusing, and then call back to it half a book later in order to be side-splittingly funny.)
> There was a scene early in one book where he talked about some plates with, very definitely, one banana on each. This was obviously significant, so I asked him to explain. But he liked to tease his audience and said he'd tell me later. We eventually got to the end of the book and I asked him again, 'Okay, Douglas, what's with the bananas?' He looked at me completely blankly. He had forgotten all about the bananas. I still ask occasionally ask him if he has remembered yet, but apparently he hasn't.
Incomplete information isn't what I think is at the heart of fantasy, and usually isn't what I think is at the heart of SF either. SF is usually a story about an idea rather than a character; it's not speculation. Fantasy is usually a heightened, exaggerated world - a world with the saturation and contrast turned right up; again, not speculation.
Speculative fiction I'd place in a niche area of hard SF that is more concerned with futurism than ideas per se. That is, it's interested in predicting how today turns out with some reasonable technological advance and specifically how the world deals with that advance, whether on a global level or via a character study.
As to needlessly multiplying entities, I agree, with caveats. It's related to Chekhov's gun: don't introduce elements that don't matter (and similarly, if you rely on something later on, ensure it's foreshadowed rather than parachuted in as a deus ex machina). OTOH, a close adherence to Chekhov's gun can make stories very predictable, for people who pay lots of attention to plots. And other people come for prose style, and still others for world-building; neither of which are overly concerned with omitting details.
The usage by modern fans and critics: http://www.sf-encyclopedia.com/entry/speculative_fiction
I also refer you to all of the review series done by James Nicoll using the term as I described - http://jamesdavisnicoll.com/
There are also 34 references for you to follow in this Wikipedia article.
Unless throwing causality out the window was the goal all along, of course.
Imagine an OpenStreetMap (it actually uses the same stack) where you can create your own fictional land - and where you can potentially interact with other people's creations too.
From their FAQ:
OpenGeofiction (OGF) is based on the OpenStreetMap software platform, which means that all map editors and other tools suitable for OpenStreetMap can be used to build OGF's fictional world. This world is set in modern times, so it doesn't have orcs or elves, but rather power plants, motorways and housing projects. But also picturesque old towns, beautiful national parks and lonely beaches.
Different users view OGF in different ways. For some, it's an exciting creative adventure, a challenging construction puzzle, or an epic tapestry of interwoven stories. For others, it's a relaxing meditation, an immersion in imagination, or perhaps the largest piece of artwork ever created — equal in size to the Earth itself. "
edits - It's not that immediately easy to find the good mapping - http://opengeofiction.net/#map=11/36.4055/119.2868 or http://opengeofiction.net/#map=15/1.6975/42.5115 is one example, and I find http://opengeofiction.net/history could be useful to see where people are working on.
This makes me wonder: has anyone tried using machine learning to generate satellite views of a fictional map like this? It seems like something that could be done -- because we have both the ground truth satellite data from Earth, and also a good representation of Earth's roads/map data from OSM, one should be able to infer, given an input OSM map (and some other data like lat/lon or proximity to coastlines), what a plausible satellite image would look like for that map.
And someone else used that data with Leaflet to make a tiled map and a globe: http://oznogon.com/golarion-tile/#6/35.642/0.000
Apropos of nothing, back during the rush to build the transcontinental railroad, the Union Pacific thought Cheyenne would become the new hub of the west and Denver would turn into a ghost town. But a quick look at a map shows the abundance of rivers and waterways flowing through Denver while Cheyenne has just a small river flowing through. Not even a railroad was enough to overcome the lack of resources around Cheyenne.
This railroading victory revitalized Denver at a very opportune time.
Moffat continued to dedicate his fortunes on continuing a railway directly west across the Rockies from Denver. He died in 1911 with his railroad companies largely bankrupt and incomplete, but in 1928 one of the successor companies opened the Moffat Tunnel, named in his memory, along the right-of-way he laid out. Now, the route from Denver west to Nevada is owned by Union Pacific, despite having been built by competitors to UP's (and CP's) route through Nevada-Ogden-Wyoming.
> And the fractal pattern of rivers is directional: rivers will always merge as they flow toward the coast, never split.
Interestingly, it is possible for rivers to split, sort of. Rivers naturally meander, and it is possible for two adjacent rivers to meander such that they intersect each other. Most of the combined flow will end up along the steeper path, true, but that doesn't mean the shallower path will dry up completely. For a real-world example, this is basically the catastrophe that Louisiana's Old River Control Structure has been staving off for decades.
Outside of deltas, it is possible, although very rare, for rivers to connect to each other and be able to transfer flow bidirectionally. The Casiquiare Canal (https://en.wikipedia.org/wiki/Casiquiare_canal) is the largest such example (despite its name, it is a natural river, not man-made). It causes part of the Orinoco River to flow into the Rio Negro and thence the Amazon.
Also, if you're into tabletop RPGs, I'm going to give a shout out to my own blog - https://wail.es/
All in all, Zach, the Dwarf Fortress creator seems to have put some thought into the (game-wise) mundane task of world-creation.
It hasn't existed for a long time, which is sad. Anyone know good (or even not so good) alternatives?
Torben's Planet Map Generator: http://topps.diku.dk/torbenm/maps.msp
donjon World Generators: https://donjon.bin.sh/scifi/world/, https://donjon.bin.sh/fantasy/world/
Experilous Worldbuilder ($8): https://experilous.com/1/store/offer/worldbuilder (previously linked to on HN: https://news.ycombinator.com/item?id=8681899 )
Space Engine: http://spaceengine.org/
Theory, code, and links:
Red Blob Games: http://www-cs-students.stanford.edu/~amitp/game-programming/..., http://www.redblobgames.com/x/1728-elevation-control/ (previously on HN: https://news.ycombinator.com/item?id=14793809 )
Procedural Content Generation Wiki: http://pcg.wikidot.com/pcg-algorithm:map-generation
Mark Rosenfelder's site is an amazing rabbithole of world-building. http://www.zompist.com/howto2.htm http://www.zompist.com/virtuver.htm
He's also written The Language Construction Kit and The Planet Construction Kit.
There's some great Google Plus communities around fantasy cartography, like Map-Making in Games https://plus.google.com/communities/105897774940532146183
I laughed. That was a fun read.
His blog is full of both granular and general advice: http://www.fantasticmaps.com/
And his article in the Kobold Guide to Worldbuilding is my go-to reference for using cartography as a worldbuilding device: https://www.amazon.com/Kobold-Guide-Worldbuilding-Wolfgang-B...
Way back in school they'd issue this in the form of a question to undergrads just to see how they'd model it / try and answer. I'm not sure if it's understood why the number is '6'
On the other hand, with how much trouble Tolkien had in general with finishing projects, if we waited for him to learn geology too, who knows if we would have gotten any books at all...
When I was a teenager reading LotR I always imagined that Weathertop was like the Tap O'Noth - it is quite a distinctly shaped hill (having a dirty great ancient fortress on top) and can be seen from a long way away.
The weird unnatural mountains around Sauron's area of the map might have been built by him using spells to create a "natural" protective wall from elves so he has enough time to forge the ring in secret.
If I remember correctly from the books (correct me if I am wrong), there is even a mention in the book about Misty Mountains being "built" by some ancient powerful beings (so that would explain ridiculous geology).
The Wall in Game of Thrones is also not natural, it was build using magic an is held together by the same spell that created it.
A retelling and continuation of parts of LotR, as told by a geologist rather than linguist. See also some of the author's commentary at http://ymarkov.livejournal.com/273409.html detailing some of Tolkien's geological mistakes.