I spent last weekend at the Oasis 21 science fiction convention in Orlando and had a wonderful time speaking on panels and chatting with the various convention goers. During a little bit of downtime between panels, I talked to one attendee about science fiction books that had been made into movies. After recounting one particularly (in her opinion) bad book-to-film translation she looked at me and asked: "Why does Hollywood do such a bad job turning science fiction books into movies?"

To which I said, "Because it doesn't matter if they do a bad job or not."

Now, this wasn't the answer she wanted, and I can sympathize. But the fact is, if you're hoping that a movie made from an SF book will be much like it, particularly a major studio movie, you will almost always be disappointed. Here are three reasons why. Note: These are not the only three reasons, just the ones we'll look at today.

1. Ninety percent of the text (or more) is going out the window.
When Jurassic Park came out in '93, I interviewed author Michael Crichton, and he said something that's on point here. It was (and I'm paraphrasing, because it was 15 years ago), "Look, when you have a novel, it's 400 pages long. When you have a script, it's the equivalent of 40 of those pages. No matter what, you're going to lose a lot." Bear in mind he was saying this about Jurassic Park, which, as books go, was about as movie-ready as they come. When you're talking about books that aren't intentionally written with one eye toward the screenplay adaptation, the challenge is even greater.

Now, Crichton's hand was on the scale a bit since much of what takes a lot of words to describe -- action scenes, scenery, character emotional responses and so on -- can be shown visually with great economy: The old "A Picture is Worth a Thousand Words" phenomenon. But at the end of the day he's correct. A lot gets thrown out, including sub-plots, asides, character development moments and (if the screenwriter is careless) key moments that fans of the book will miss and howl over.

2. The definition of "hit" is vastly different for books than films.
A science fiction or fantasy book that sells 100,000 copies is, these days, a big hit. A major science fiction or fantasy movie that had 100,000 theater admissions, even just on opening weekend, would be a miserable failure; that would be the equivalent of a $700,000 box office gross. Not to keep harping on poor Speed Racer, but its $18.5 million opening weekend means about 2.64 million people actually shelled out $7 to see it ($7 being roughly the same cost as a paperback book). If one of my books sold 2.64 million copies -- ever, forget just in its opening week -- publishers would drag themselves across iodine-saturated glass shards to have me be in their book lines. And I, for one, would love to see that.

The reason the definitions of success are different are pretty simple: The number of people required to produce and distribute a book, and the cost of production and distribution therein, is several orders of magnitude smaller than either number of people required to produce and distribute major films, or the cost to do so. This means that books can afford not to take into consideration the lowest common audience member. Whereas if you're making a film -- especially one with lots of expensive effects, as science fiction films so often have -- you have to try to get as many people as possible into the theater from the word go. What that means for your favorite book is that all the parts that really speak to a core fandom are likely (to mix metaphors here) to get flattened out and homogenized and made as blandly acceptable as possible. Because ultimately, millions and millions of dollars are on the line. Art and textual fidelity are nice, but having big fat hits that spawn sequels and keep Hollywood afloat are nicer, from the film company point of view.

3. Textual fidelity doesn't have much of a track record in science fiction film.
Here, courtesy of Box Office Mojo, is a list of the most successful science fiction films based on books since 1980 (fantasy is not represented). What do we learn? First, we learn it helps to have Steven Spielberg direct, since he's in the top three spots. Second, we learn that fidelity to the book text has almost no correlation to the success of the film. Jurassic Park and The Lost World are somewhat related to their books, but War of the Worlds is a largely complete re-imagining, and if you want to see science fiction literature fans get truly foamy, ask them their opinions about the adaptations of I, Robot and Starship Troopers (remember to hold up a tarp to shield yourself from the spittle).

Basically, you can't make the argument to a film studio that it matters if the film is only tangentially related to the book, because the ultimate argument -- that messing with the story messes with the box office -- isn't there. Perhaps as art it might make a difference, but on that score it's worth remembering that since 1980, the most artistically influential SF film adapted from a book was Blade Runner, which is in fact not very much like its source material, Philip K. Dick's Do Androids Dream of Electric Sheep? at all -- and which was a bit of a financial failure when it first came out.

What do we learn? That when a book and film share the same title, it doesn't mean they share much of anything else. Nor is that really likely to change.

Winner of the John W. Campbell Award for Best New Writer, John Scalzi is the author of The Rough Guide to Sci-Fi Movies as well as the novels Old Man's War and the upcoming Zoe's Tale. His column appears every Thursday.