Re: Worldbuilding and Infodumps
In response to a discussion on worldbuilding and exposition in tsubaki_ny 's journal, because I got longwinded and it wouldn't fit in a comment box.
Some thoughts on the subject from an avid reader and consumer of a myriad forms of fictional entertainment:
It's important to not allow exposition to turn into an infodump, of course, because it causes the story itself to grind to a halt. It's also important to not leave your audience confused because they lack vital context about the setting.
Taking a look at the example of Tolkien for a moment... yeah, we love the man for essentially creating the modern fantasy genre. And he made an incredibly detailed world to set his stories in. But he wasn't very discriminating about what information about that setting was actually relevant to the story he was currently telling and what wasn't. I read The Children of Hurin last month, and in the early chapters I had a devil of a time getting into it, because it's quite a while before we even meet our protagonists: instead we get mounds of backstory on noble families and elves that won't really be very relevant to the story itself. (By contrast, one of the most important battles in the history of Middle-Earth, an extremely dramatic event which also defines the course of the life of our main characters for years to come, only gets the vaguest of recaps.) Tolkein, I think, wrote to try and emulate the style of ancient epics. In the epic poem of Beowulf, for instance, at the end when Beowulf is being killed by a dragon and one of the other warriors with him, Wiglaf, draws his sword to go help, we're then treated to a completely unnecessary relaying of the backstory of said sword, all of which is not really relevant to the story of Beowulf's death. (He also takes 28 lines to essentially tell the other warriors, "You cowards! I'm going to go help him!" the entire time of which I would imagine Beowulf is muttering, "Hey, I'm being killed here! Jumping in to help sometime this week would be nice!") While all this extra information might be handy for the keeping of an oral history, especially since Wiglaf got to be the next king (there's no excuse for the absurdly long speech, however... it takes him longer to say "I'm going to help" than most fights last... unless, of course, he was trying to get Beowulf killed so he could replace him...), it's not really very good for the narrative flow of the story or very important to the story's subject matter, namely Beowulf himself.
Worldbuilding information can be valuable, but you need to finesse it in there so it doesn't disrupt the flow of the story, and you need to know what is important information that's needed to read the story and what can stay in your head or your private notes until being included in an appendix/technical-manual/side-story/RPG-setting-guide/internet-forum-post-to-settle-a-debate. It's not so much that nobody cares: we're nerds, after all, and there will always be somebody who's just absolutely dying to know. But even if everyone cares, it's still breaking the flow of the narrative, knocking the reader out of their experience of enjoying the story itself.
And while you may or may not agree with me on this, I still hold firm to this opinion: the primary goal of a novel, or any work of narrative fiction, should be to tell a good story. Worldbuilding should be at best a secondary concern. A setting can be valuable to a story, but it is not in and of itself enough to prop up a story.
For example, let's examine the original Star Wars trilogy. Films and not novels, I know, but examples with which we are all familiar. The standards for exposition and infodump are different in different mediums, but it's a still a useful point of comparison. Pretend that we're discussing the novelizations, if it helps. The body of supplemental material regarding the Star Wars setting is massive, far in excess of anything Tolkien created. (For comparison purposes, the Star Wars wiki contains more than 75,000 articles, to the Tolkien wiki's paltry 3,600. If there's a single setting with a larger body of information about it, I will eat my hat.) Only a tiny fraction of this information is actually in the films, however. Admittedly, much of it was created after the films (an enormous portion of the setting was actually created by the Star Wars Role Playing Game in the 1980s), but even if the information had existed, it would not have been included, and much of what information did exist at the time was not included: it wasn't crucial to the story being told. It might well provide valuable character insight to know that Chewbacca is fiercely loyal to Han Solo because Han saved him from Imperial slavers, and that doing so cost Han his hard-earned commission as an officer in the Imperial Navy, but the story can continue on without knowing that, and even if that information were to be given to the audience mid-story, an author would would be wise to find an excuse for it to come up in conversation (in as unexpository and character-appropriate a way as possible) and not bring the cantina scene to a grinding halt to narrate it to the reader, as Tolkien likely would have done if by some bizarre happenstance he were to write a novelization of the film. Extending this to the absurd, it is absolutely true that many of the people and aliens present in the cantina in that same scene have had names and backstories established in later works, but that doesn't meant that the audience needs to know them. It might well be that the stories of those people are interesting in their own right (more so than the story of Wiglaf's sword, at least), but they're not relevant to the story of A New Hope. To follow the story of Luke and Han and Leia taking on the Empire, we don't really need to know that the cantina band are actually big-name performers who are stuck there because they lost their instruments in a card game, or that the guy with the horns is a fugitive war criminal obsessed with music who cheated at cards to get said instruments for that purpose, or that the guy who looks like a werewolf meets the love of his life at the cantina that day. Those are somebody else's stories, that can wait to be told in the Tales from the Mos Eisley Cantina anthology. (No, I'm not joking. It's a real book.)
Whereas to prove that having an interesting setting won't make up for boring exposition, awkward dialogue, unlikeable protagonists, poor plotting, and inconsistent characterization, one need only look at the Star Wars prequel trilogy.
From the other direction, to see what happens when you make a novel more about worldbuilding than story, we need only to look at a writer named Ed Greenwood and a setting called "Forgotten Realms". In 1987, Greenwood published the Forgotten Realms Campaign Set, a setting guide for the tabletop RPG Dungeons & Dragons for a fictional world that Greenwood had been developing as a hobby for the last twenty years. There was certainly a wealth of worldbuilding information present, as there should be: the purpose of an RPG setting guide is to allow the readers to create their own stories and characters within that setting. Unlike a novel, worldbuilding is the primary purpose. Forgotten Realms has proven to be one of the most popular D&D settings ever to this day, not only for players to set their own stories in, but for a slew of novels, video games and comic books of varying quality.
Unfortunately, some of those novels were written by Ed Greenwood himself, and I've yet to read a novel of his that I actually enjoyed. The first one, Spellfire, was certainly loaded with worldbuilding info. We're introduced to Myth Drannor and the Knights thereof, dracoliches and the Cult of the Dragon, Elminster, the Simbul, Zentil Keep, Manshoon, Silverymoon, Mystra, Spellfire and a bunch of other elements of the Forgotten Realms setting... and most of them are either unnecessary or completely tangential to the story. A subplot about a power struggle inside an evil magocracy has no bearing whatsoever on our protagonists, and doesn't amount to much. The final battle is against an adversary with whom our leads have no prior history. And the main characters are all rather flat and archetypical and seem mostly to be motivated by following the demands of the plot. Greenwood was more interested in introducing his world than in telling a good story. (The other book of his that I've read, on the other hand, I didn't like because it's all about his Gary Stu of a pet character, Elminster, who I find boring as hell.)
Alternately, for an example of how even an RPG book--where the primary purpose is worldbuilding--can try to make its exposition read like part of a narrative rather than pure infodump, one can look at the older sourcebooks for the cyberpunk/fantasy game Shadowrun, in which setting information would often be presented in the form of internet posts, annotated by other posters' comments containing their own experiences with the subject, alternate viewpoints, additional information in the form of rumors and innuendo, accusations of bias, arguments or even foul-mouthed trolling.
The key to working worldbuilding into your story, I think, is to remember that the story comes first and the worldbuilding is not so much icing on the cake as carbon in the alloy: an indistinguishable, inseperable part of the material that makes the base element stronger.