Sarah Hoyt, author and blogger has an excellent post on the differences between editing and proofing.
A critical read for every aspiring author.
Basically proofing is correcting typos, common grammar mistakes and the like. It's a facial and a new hair do. Editing is major cosmetic surgery. A good editor knows how far or how deep to cut and where it has to be stitched back together.
I have had the novel "finished" for some time now, in the sense that the story is over, the plot resolved, the characters have traveled their emotional arc, but like building furniture, all the pieces may be together but it isn't finished yet. It needs another fourteen layers of lacquer and polish, which usually takes as much or more time as building the darn thing in the first place. And often you have to wait for long times to let one coat of varnish dry and set before the next can be applied. In writing you have to wait for things to gel and set before you can tell what's wrong with it. Errors that you are oblivious to today become glaringly obvious six months down the road, as any voter disappointed in a candidate understands.
This is a soul-killing process and very frustrating.
Your characters are very much your family and making major alterations to their lives feels like going to your own living children and saying, "You're too short...grow up some more!" Characters are like living organic things, they don't want to take orders or change, but sometimes you have to insist, and sometimes you have to send them to fictional nirvana by removing them from existence.
So you labor for months to fix what took years to write, or worse, dither endlessly uncertain of what to cut or what to change. I am currently in what my wife calls "hand-slapping" stage. That's the stage where you have to slap the hand of your children to keep them from picking at a scab. Editing is as much about knowing when to stop as when to start.
There's nothing wrong with the novel, but I can't resist the urge to tinker, but eventually you tinker so much in one area you have to open whole other plot lines that have long been settled. So you end up changing stuff that need not be changed, until you've rewritten the whole thing again, which of course, demands another rigorous editing session. Lather. Rinse. Repeat until you put your head in an oven. This is why we set a firm release date for the book, to prevent this kind of work-crushing self-kibbitzing. Set it and forget it after that.
But maybe there is a better way.
Here's where I think e-publishing has an answer...kinda.
In dead tree publishing the thing has to be "set," at one point in time. This was once literal. A typesetter had to set the type before production. This means that editing ends when the typesetter starts his craft and begins loading those tiny little metal metal mirror letters into the tray. The technology changed over the years but it was still the same, nothing could be printed until the text was finished. In the digital age however, the editing can continue even after the publishing.
On Kindle, I can update my work, trim it, and fix the embarrassing typos should any slip in, but I can also continue to edit story, ideas, characters, updating them constantly if I so wish long after it's been published.
If the book isn't selling, or if I get comments that one chapter drags, I can edit it. If they think the cover really doesn't tell what's inside, I can change it. I enter into a dialogue with my customers about the final product, improving it with each round. I used to work in marketing and this is exactly how the process worked in advertising. If an ad wasn't selling, we didn't wait for it to find an audience. We killed it and moved on, changing or updating the sell until it...well...sold. The same could apply to fiction.
The person who bought my book this week may get a slightly different book than the guy who bought one last week. The newer book would be presumably better. We are already used to lots of products being updated, why not fiction? We could see a future of endless feedback, using the crowd to beta test our work. Why have two or three first readers when you could potentially crowd source it and have hundreds? You might even release the work rough at a lower price to encourage these early adopters. Then take their advice and put on a few more coats of varnish later.
(Bad news for freelance editors if this takes off, but hey, I used to work in slide libraries as a visual resource consultant, a job that doesn't exist anymore thanks to digital images, so dem's the breaks kid.)
In theory, a work might never actually be "done" in the traditional sense, but might be constantly evolving.
At some point this would probably be pretty ridiculous. I'm sure that an endless series of edits, though it might thrill some avant garde types, is a pretty good recipe for a lousy book. However, maybe we take on too much control and need to trust and grow with our fans. All authors do this, but they do it over the course of several books. Now with e-publishing, they can do it over a single book!
I'm not advocating putting up just any rickety piece of crap. An author would ruin his reputation that way, but if you do solid work, then why not let them see you put the final touches on it?
At some point the author must assert control, but right now I'm looking at this piece of furniture. It feels solid and smooth, and sanded. Nothing wobbles, and I'm thinking I'm going to skip the 18 layers of french polish and just rub it down with linseed oil and send it on to the customer and see how they like it. I can always add a touch of varnish later if they want it.