This is currently undergoing fitful development on the plates of director David Fincher and star/producer Morgan Freeman. I’d give it a one-in-twenty shot of actually going into preproduction.
A question for everybody who has responded so far:
I’m seeing a whole lot of suggestions for adaptations of material that, at least on the page, are basically unfilmable. However, there are a number of excellent story ideas that could be adapted, if you’re willing to consider how the rules by which movies operate are different from the rules by which prose stories work, not to mention basic pragmatic details of the (sometimes now-dated) original material.
Just one example:
Heinlein’s The Moon is a Harsh Mistress is a good suggestion. The sociopolitical explorations that have always been Heinlein’s bread-and-butter are nicely balanced with the fascination of the locale and the ample action scenes. You can be essentially faithful to Heinlein’s themes about economics and human group dynamics while still delivering the effects and slam-bang stuff that is required for a science-fiction movie to get greenlighted in Hollywood. Plus, the intelligent computer Mike (or Michelle :)) is a fun character, and something the movies haven’t really done before: superintelligent but persnickety, halfway between HAL and the bomb in Dark Star.
But: Heinlein gives his characters an offbeat linguistic style, a patois of choppy English and a bit of Russian that makes sense from the perspective of the mid-1960s when the book was written. Now, though, the Russians are fading in international prominence, and the dialect looks like a Cold War relic. You simply couldn’t pull it off in a modern film; it’d make the movie feel like a museum piece.
So you have two choices: You can dump it entirely, or you can change it to something more realistic. Based on current trends, it seems to me that the most likely candidates for non-Western integration into future space/engineering projects are China and India. If you want to keep Heinlein’s general idea, which is quite reasonable, you’d need to come up with a new variation on his original notion, one that incorporates Chinese (as in the Firefly universe) or Hindi. Or you could make up something else entirely, projecting a random future event that makes, say, Egypt a major power.
So as we fantasize about the movies we’d like to see, I need to ask: How reasonable and realistic are you about what would need to be done to translate the written work to the screen? Consider that the faithfulness of Peter Jackson’s LOTR trilogy is viewed through two lenses: the purists who hate the exclusion of Glorfindel and a thousand other details, and the casual reader who couldn’t care less and who has gotten the basic gist of the story. Me, I recognize Jackson’s films are not perfectly “accurate,” but I also recognize that they’re far, far more faithful to the source material than anybody could possibly have expected given the magnitude of Tolkein’s creation. It is a necessary compromise. (And no, I don’t want to argue whether the specific choices were good or bad. The fact the remains that choices had to be made.)
So stop and really think about it. If you’re going to commit, say, Asimov’s Foundation material (in some fashion) to the screen, how do you do it? How do you beef up his notoriously thin characters? There isn’t enough in the text to make interesting people, so you have to add stuff; what do you add? What is the story arc for two hours? Who is the primary individual protagonist? Who is the primary individual antagonist? What is the central thematic point of the book(s)? What is most important to bring to the screen, and what can be discarded? If you choose to focus on the Mule, how do you deal with the fact that a lot of his action occurs offstage? How do you stretch a story over hundreds of years, regularly replacing your cast of characters (excluding the Seldon recordings), without losing the audience’s interest?
As another example of a story whose structure just plain wouldn’t work on screen, consider Haldeman’s Forever War. There’s no central dramatic question, and no narrative arc. The power of the story is in the character’s deteriorating relationship to his job, and in the accumulating evidence of the passing centuries of history that the leadership, despite ever-advancing technological wonders, continues to have approximately the same lack of clue about what they’re doing as they always have. Great book. Would make a terrible movie. If you’re going to adapt it for the screen, you need to find a central question to drive everything, probably something about separating the hero from his lover toward the beginning, and having him assume the relativistic dilations mean they’ll never see each other, but then give him a hint at a possible reunion if he does this-and-that-and-the-other. In other words, you’d basically have to make up the last twenty minutes of the movie completely from scratch in order to have a screen story that works. Does it violate the book? Arguably. Is it necessary? Yes.
I hate to be a grouch, but translating a prose story into a visual medium is a lot harder than it seems. It’s one thing to wish for a perfect adaptation to arrive as if delivered from the sky, but it’s another to actually have to figure out how to slog through the process of making it happen.
Speaking for myself, I think the simplest-to-summarize ideas are probably the most potent for screen treatment. Take, for example, Robert L. Sawyer’s Calculating God. The hook is fantastic: A spaceship lands next to a museum. An alien creature gets out, walks in the front door, and says to the receptionist, “Take me to your paleontologist.” I mean, how can you not love that? The ad campaign writes itself, and as long as the rest of the story bears a passing resemblance to the book, you’ve got a winner. Anything more complicated, and you’re in trouble; compare Sawyer’s Factoring Humanity, which while based on a cool idea, the premise would take more than twenty-five words to explain (and is more than a little reminiscent of Contact), and thus is dead on arrival in movie terms.
The bottom line: In a movie, you’ve got 45 to 60 individual scenes, and maybe half that many important plot points before the story gets too complicated to follow. You get five minutes of pure exposition at the beginning before the audience gets restless; if you can’t explain your science or your civilization in that time, you’re dead. You get two or three major good-guy characters, four or five major supporting characters in their orbit, and half that many on the antagonist’s side. Within the first half hour, you must establish a central driving narrative question that will be answered in the last five minutes (why is the monolith on the moon? will NASA divert the comet? will Cole track down the virus?).
I’m not trying to piss in the fire here. I am asking that we intelligently and creatively consider what would need to be done to help the story leap successfully from page to celluloid. I’m all for good SF movies; I’m regularly disappointed by Hollywood’s pale imitations, and absolutely encourage the adaptation of worthwhile stories in a responsible manner. But just as you must be responsible to the source material, you also have to be responsible to the requirements of screen storytelling. Otherwise you get David Lynch’s Dune, which looks utterly fantastic and which is so stilted and awkward from a narrative standpoint that it’s utterly impossible to sit through in one go.
Back to your thread in progress. 
