First off, we need to get our terms straight. When I say Science fiction, I am not, repeat, not talking about Star Trek: The Next Generation.
I'm talking about Heinlein, Bradbury, and Herbert. I'm also talking about Asimov, Clarke, and Niven.
See, there are two classes of Science Fiction, and one of them is terribly misnamed.
First, there's "hard" science fiction. In hard sci fi, the science matters. Usually, the science, technology, or concept under discussion in intrinsic to the plot. For a good deal of this particular grouping, the characters are completely unnecessary, and the writing shows it. As an example, if you read I, Robot (and I'm not talking about the movie), I would ask you to describe three characters from the novel in more than one sentence. Hard to do, isn't it? That's because, by and large, they don't matter. What Asimov does, and does exceptionally well, is present us with the three laws of robotics, and basically walk through how they can subverted.
Hard sci fi has varying levels of character development, granted. I had a good feel for Heywood Floyd in 2010: Odyessey Two (although, to be honest, a lot of that came from Roy Scheider's portrayal in the movie), it was again the science that grabbed me; I still find myself hoping, even if I'll never live to see the result, that Jupiter is indeed harboring a diamond larger than the Earth in it's core.
In hard sci fi, the science matter. In good hard sci fi, you learn something. In really good hard sci fi, you keep thinking about the subject long afterwards.
In bad hard si fi, the science matters, but it isn't science. For example, how many times in Star Trek TNG did deus ex machine rear it's head in the form of "reversing the polarity" on some completely random peice of ship equipment? Too damn many, at any rate.
In science fiction, not the hard kind, on the other hand, it's the people that matter.
I had a conversation with my son a while back. He's seven, and he wanted to know why I had such a high opinion of Shakespeare.
"Because," I said, "unlike most of his contemporaries, he's still relevant. You can read Shakespeare, and what he has to say about people is still true. You can replace the castles and kings with plantations and aristocratic genteel southerners, or modern folk, or spaceships, and the story will still work."
I know this to be true, because with Shakespeare, it's been done, and it worked. For example, check out Ian McKellan's recent version of Richard the Third; even set in 1940's with Richard mirroring Hitler, the story rings true, and works.
This is also true of good science fiction. For example, Starship Troopers by Heinlein, which to this day I hold as one of the greatest examples of it's type ever written, could be easily moved into a World War II setting, or Victorian England, and still work. (I am not referring to the movie; Paul Verhoeven earned himself a special place in hell for committing an act of desecration when he directed that abomination.)
Dune, by Frank Herbert, is another great example, although part of the reason it works is because it's already a transplant. Dune is a stream of conscious work (if the thinker is a genius) on the modern world's relationship with oil.
Science fiction, terribly misnamed, is really about people. Good science fiction resonates with us no matter what the setting. Great science fiction changes how we perceive people.
Bad science fiction say all the wrong things about people. Quite a bit of bad science fiction focuses far too much on our baser aspects; really bad science fiction is driven by sex that it might as well be a Harlequin romance novel.
The genre, in it's literary format, is dying.
It was perhaps inevitable; our world changes so much, technologically, that instead of being prescient (like Arthur C. Clarke predicting cell phones and satellites) to being outdated the day after it's published. Being a good futurist these days is a difficult proposition, since there's so much change, and also so much uncertainty. A lot of authors are playing it safe lately, and focusing on Gold farming and MMOs. I can't wait for that particular fetish to die.
There are a few authors out there still working, still putting out quality work. I really enjoy Neal Stephenson's work (even though even he's caught the gold farmer bug); I think Anathem is also potentially one of the greatest books ever written. (That said, I would strongly encourage you to catch the audio book; in addition to excellent writing, it's one of the best readings I've ever heard. The reader (who's name I cannot find, sadly) does an incredible job of performing.)
Still, the days of Bradbury are far behind us, and I mourn their passage. We've lost a cultural sense of innocence, and with it went a sense of awe, of wonder, and joy that I find myself rereading older books to regain.
I'm working up a list of book recommendations; these are the thoughts leading up to that list, which is coming soon.