(Sorry 'bout the delay: I've been really busy.)
I have a confession to make: I am not a gamer. No, really and truly. My first computer was a Sinclair ZX81, which was so crap that I sold it after two months and bought a Casio FX-702p — a glorified programmable calculator, but I still have it and it still works. I lived computerless until 1985, when I bought an Amstrad PCW8256, and went straight from there to PCs. My computing trajectory therefore missed out the Commodore 64, the Spectrum, and then the 16-bit machines — Atari ST, Amiga, and Archimedes — and then diverged into the UNIX/Linux world. By the time my work machines were able to run games reasonably well, my eyesight and reflexes were those of a thirty-something. Lacking years of twitching (or rather, having keyboard reflexes fine-tuned for vi, not missile launchers), I am consequently crap at almost all computer games.
I've been really interested in Virtual Reality as a key indicator that we're living in the future ever since William Gibson published Neuromancer back in 1984, and especially since Neal Stephenson published Snow Crash in 1992, and rang the curtain down on cyberpunk in the process. (Yes, I have first edition copies of both books. Well-read ones.) VR seems to exemplify the modern SF paradigm of information space in much the way that the idea of interplanetary exploration exemplified the old paradigm of real space. But, here in the real world, the progress of VR into everyday life has been slow and tenuous ...
Then in 2004 I was in the audience at a panel at an SF convention discussing crime and the internet. One of the panelists had an anecdote from a police station blotter. "Man walks into a police station in South London early last year. 'I want to report a crime,' he says to the desk sergeant. 'I bought a magic sword, only it's not enchanted correctly.'" It took a while for the perplexed officer to understand what was going on, but it was indeed logged as a crime: fraud. (The victim had been sold a duff in-game item via an auction site: offense, "obtaining money (or goods in kind) by deception".) This caught my imagination and I went digging and stubbed my toe on the economy of Ultima Online (paper from 1999), and then the whole then-embryonic field of MMO economics. That you could buy and sell items using real-world currency seemed inevitable, and the prospect of an inflationary production-driven economy sucking in more and more players seemed fascinating. As a friend put it, "MMOs like World of Warcraft are the first commercially successful Virtual Realities to get more than a million users".
This was around the time of the Second Life boom (yes, I dabbled: yes, SL is still going), and all in all it suddenly looked as if, after being fiction for a decade, and then hanging fire due to GPU performance issues and motion sensors not being up to snuff for another decade, it was finally coming through. This was also the point at which I had my first smartphone (Palm Treo 600). So in a fit of creativity I realized that I needed to write a novel about virtual reality — not the by-now cliched cyberspace of the 1980s and 1990s, but the real thing.
Because it would centre on a robbery at a gaming company that threatened to bring down a real world economy, it'd be told in the second person — the natural voice of the classic text adventure game ("you are in a maze of twisty little passages, all alike"). It would be set daringly close to the present day, just a decade ahead; 90% of the world would be the familiar already-here universe of my home town: 9% would be the predictably-there stuff of the proximate future — take Intel or ARM's road maps for the next three years today and three years ago: draw a straight line extrapolation to get to where our capabilities will be in a decade (yes, I know that's a naive approach, but it works for fiction): and 1% bugfuck strange, alien, and as unexpected as LOLcat videos back in 2003. In addition I assumed everyone would be using 3G or 4G mobile broadband and wearing their smart lifelogging device/mobile office on their head (as BT's Peter Cochrane was doing in the late 1990s): and with ubiquitous GPS integration and imaging, it'd be possible to map a visual representation of the internet onto the real world. (Back in the 1990s HP research were looking into defining an HTTP-like protocol for retrieving information from the internet based on the GPS location of the requester; I assumed a distributed, dynamic DNS-like protocol for servicing such requests would be available by 2015. How optimistic of me: in the early 00s I didn't anticipate HP's reign of misrule, or the downsizing of corporate research departments!)
Finally, it would be written as Mundane science fiction (i.e. no ray guns or aliens, just believable technological and social change). I allowed myself one borderline-implausible device in the whole novel — the quantum computer used for breaking PKE.
Anyway. I went to pitch this to my editors at Ace and Orbit.
At Ace, my editor, an old veteran, gave me a long, cool, appraising stare then said "I'll take it, on condition it's part of a two book contract, and the other title is a space opera." (Translation: "if your wild-ass experiment fails, the sales of the space opera will cover it.")
Orbit in the UK ... had just been sold by Time-Warner to Hachette. Everyone there was extremely nervous at the time I made my pitch, wondering if an axe was about to fall. As it happened, the axe was imaginary; but my editor didn't feel able to gamble on an experiment — he offered an advance half the size I was expecting, direct to paperback for a short run.
My agent and I turned him down, and agreed that I'd take the risk of writing it with just a US sale, and see if Orbit would take the finished manuscript when they saw it. This seemed like a big risk at the time — it knocked 25% out from under my bottom line. It seemed even riskier when the novel took me longer than expected to write: I turned it in three months late, sweating bullets, after taking 15 months over it. (That's three months during which the author doesn't get paid. It tends to concentrate the mind!) But, finally, we got to show the finished work to Orbit ... and my agent got a more-than satisfactory bump in the advance they were offering.
Final note: "Halting State" was something like my tenth SF book in print. And it was that utterly unpredictable thing, the breakthrough novel that makes a career. It ran away. Ace upped my hardcover print run, then sold out before publication and had to reprint. Starred reviews, Hugo shortlist, big paperback print run: it earned out its entire advance in the first month out in hardcover, and six months later a large (and unexpected) royalty cheque landed in my bank account. It did well for Orbit, too — in their scramble to inject it back into their schedule they didn't have time to run a hardcover release, but it came out in trade paperback and kept being mis-filed under Crime in most of the bookstores I looked in. Interesting fact: Crime fiction outsells SF and Fantasy combined by about 3:1. (If I was purely focused on making money by writing fiction, that's where I'd be. (But one word of caution: writing crime is a lot harder and more specialised than I realized before I tried it. It's got the research requirements of hardcore physics-driven hard SF, combined with the characterization requirements of literary fiction, and your plotting had better be bulletproof or it just won't work. (Hint: write the first 90% of the book, with lots of random clues and characters. Then go back, re-read it, decide who Did It, delete contradictory evidence, and write the end.)))
Questions? (I'm a bit busy at present but I'll try to set some time aside for answering them over the next week.)