This topic is by no means complete. There is so much more to delve into just about starting a novel, but I think it is time to clean this topic up a little and move on to another. So I'll finish with this. First, the four basic rules I employ when writing:
1. Don't confuse your readers.
2. Entertain your readers.
3. Ground your readers in the writing.
4. Don't show (or tell) everything.
The last rule has a double implication--it implies "show don't tell." This is, of course, the basic rule of writing--show don't tell. Arlo Guthrie's book, A Field Guide to Fiction Writing, is easily the writer's manual on this topic. I won't go into anymore details today.
The second point that I haven't touched on yet is the idea itself. If you want to write a novel, you need a novel length idea--an inspiration. No cheating. Although it can be a good exercise, it doesn't do to take your favorite author, movie, program, novel, or anything else and write a similar novel. It is silly to think you can be successful by copying any work and producing something similar. To be your novel, the ideas have to be novel. The problems with many works today is they are too similar to other works, they have a voice that is too like other works, or they are so trivial, they are meaningless. That's the other side of the coin. A work that has a trivial theme, plot, or storyline is not going to be successful and will likely not get published in the first place. The question of what is "trivial" is a difficult one. I can tell you that if you have more than one unique miraculous incident in your novel, it is likely trivial. We call these a deus ex machina, a god machine. A novel can spin from a singular unique setting or event, but when they keep happening, the work's credibility becomes a question. For example, although I don't buy it and few readers will (unless you absolutely build it properly) love at first sight is a cliched and deus ex moment. A novel based on "love at first sight" might have a chance in the hands of an absolute expert writer, but the theme is so old and has been used so much that no one will believe it today. Further, if the theme and plot of such a novel isn't "love at first sight," any other unique event that propels the novel will become trite and unbelievable. Your novel can't become trite. Unique events can flow from one another in the novel based on the singular circumstances developed in the novel, but those must come logically from the others.
Example from Dana-ana http://www.goddessnovel.com/. Dana-ana is about a girl who acts like an Anglo-Saxon maiden although she lives in the modern world. The plot is basically wrapped in a question: who really is Dana-ana? It is a revelation novel. That is, it reveals more and more through the characters to the point where everyone, at the end, finally gets just who is Dana-ana. The storyline develops around this revelation. The unique point of the book is Dana-ana herself--everything in the novel flows from this simple evidence. Everything becomes explicitly concrete because of who Dana-ana is. The reader is grounded and funnelled into the story because of this singular point. Further, the theme is about redemption. It is the redemption of Dana-ana herself and the redemption of those around her. Why does Dana-ana need redemption? That is bound up in who Dana-ana really is. So, you can see, a single unique existence propels the novel. There are magic and somewhat miraculous happenings in Dana-ana, but it is a fantasy-like novel with a fantastic plot. Those magic and miraculous happenings are not unique in the structure of the novel--just as future technology is not unique in a science fiction novel. When the basis is future technology or magic, the events are no longer unique, they are common, but that is a topic worth more discussion.
I hope this short 10 part explanation of how I start to write a novel was helpful. I think I'll talk more about inspiration tomorrow. Until then...