I don’t know why I expected Dan Simmons’ Ilium to suck, but it most definitely does not. Maybe it’s my near utter disinterest in the whole Iliad thing; Gods and heros acting childish and stupid, jabbing swords into each other on some godforsaken plain in order to settle a custody battle over some king’s tart. But Simmons breathes some life into the whole affair, giving you characters that you might care about instead of a mob of unlikeable, braggart, numbskull heros. Recommended, even if lists of Greek and Trojan casualties give you hives. (The concluding book is out, so you don’t have to worry about getting Farmered).
Most CS writing makes me want to pluck my eyeballs out with a spork. Joel Sposky has a collection of essays by other people out, The Best Software Writing I, and it looks pretty refreshing. Most of the contents you can already find on the web (e.g., Your Starbucks Doesn’t Use Two Phase Commit), but it’s nice to have some decently written material all in one spot. It’s a nice, light read when you’re not actually after, say, how to make a two-phase commit actually work.
I have a love-hate relationship with Charles Stross’ Accelerando. It’s a number of short stories glued together into a novel, following a rather screwed up family through three generations as they live through a Vingian singularity.
Singularity? Briefly: A point in time where the curve of acclerating technological advances becomes so rapid (e.g., a week, a day, an hour in which advances are made that dwarf all previous periods) that you can’t make any predictions about what will happen. Some givens: Way too powerful sentient hardware, working nano-technology, economic systems that are basically too advanced for humans to understand or participate in. You are not expected to be able to understand what is going on. Most singularity stories deal with the smoking rubble after the curve melts down.
The problem is that it is extremely difficult to write about characters that are smarter than you are. Niven noted this when he wrote Protector, and sweated tons of details that only become apparent long after you finish the book, and John W. Campbell had a policy of rejecting stories about supermen (One rejection slip to a famous author: “You can’t write this story. No one can.”) It’s hard to know what a super-intelligent being will decide what to do.
Stross makes it palatable by following characters who are (mostly) just ordinary humans, maybe a little smarter than we are, or disadvantaged in certain ways (e.g., slightly crazy). He uses the inevitable super-intelligences as a backdrop, like a threatening sea-storm that could take out the little boat. Protagonists in Accelerando are reduced to “rats in the walls” status, avoiding contact with the big, bad smart things. But even handled with kid gloves, the book starts reading a little like Robert Reed’s Sister Alice, which unfortunately does attempt to deal directly with characters that have superhuman ability and intelligence, and winds up just about as boring as any of E. E. Doc Smith’s planet busters. You’re always left wondering what whizzbang gadget the author is going to whip up to get the damsel out of distress, because anything goes when you know more than the reader.
Don’t get me wrong: I liked Accelerando a lot, especially the last chapter, and I do recommend it. It’s just somehow unsatisfying in terms of what the characters can do about their predicaments.
And I do want a pair of those magic heads-up-on-steroids glasses that the main character loses in the second chapter . . . after a suitable low-level format.