angeyja was asking about Sampo on e-mail, and my response ties in with how this livejournal has evolved: I was journaling about writing these books back when it was supposedly going to be "this book," over on novel_gazing. Most of the people who were initially on my friendslist here knew me from NG, but the rest of you have kind of filtered in a bit at a time.
So here's why I'm doing such huge revisions to Sampo: I was writing this book for the better part of three years. Granted, I wrote three other books in that time, two YAs and Thermionic Night. Still. When I started, I had very vague ideas of how the whole thing would be shaped. If I had been right, it would have been awful. It would have been more like an outline than a book, and nobody who didn't already know a lot about early computing and Finnish history and mythology would have understood a word of it. (Yes, I was on the way to writing a book only Linus Torvalds could read. I don't even know if he'd have enjoyed it.)
When I figured this all out -- when I figured out what a mistake it would be to write it that way -- I had already divided the thing into sections, so I renamed the section files and went on drafting until Thermionic Night and Sampo had words on all the bits of them. (Midnight Sun Rising is part of the same thing but also stands alone, so I have let it sit by itself for awhile at 40K or so, while I mess with the others.)
So because I don't write chronologically, there are scenes in Sampo that were written as early as anything in Thermionic Night. Which means that I didn't have all my ducks in the proverbial row; I wrote things that fit much better with my previous concept of how the whole thing would go, or that didn't fit with anything at all, really. I seem to need to get things wrong before I can get them right, sometimes, and I have had to make my little perfectionist peace with that.
As a result, the revisions to Sampo are...large. Yes. They are large enough that I wouldn't give it in its absolute rough draft form to timprov and markgritter, because even they could not make sense of what on earth I was getting at there in Chapter 7 and where did Lucy go and where did Aki-Petteri come from and are you sure we have to care? Large, large, large. They are also easy, because I know what I'm doing now, so I can see what I was groping at before. Edward had a gun in one of the early chapters of TN, and while I didn't put it on the mantelpiece, I did bring it out: looky there, Pa, a firearm. I even knew who he shot (was going to shoot?) and when. I was way off base with why, but only in the sentences directly surrounding it. The underlying bits of the real reasons why are all there for even a child to see, providing that the child could get through all the bits about Chapter 7, Lucy, etc. I haven't even gotten to the shooty bit yet, and it's clear to me what I did wrong and how it has to go right.
This is why this is the cheery bit of revision: because it's a matter of putting things right. Because I know not only what I screwed up but how to fix it both well and quickly. So I'm able to be kind to the self that wrote the earlier draft, because now she looks like a two-year-old trying to tie her shoelaces: the goal is clear, the problems are clear, and she hasn't got it far enough wrong that I can't take the shoes off, put them on the other feet, tie them neatly, and head off to the park. Later we may find that she's outgrown the shoes, that the dog chewed through one of the laces, that there's a funny spot where she's worn through the shoe using it as a brake with her wagon. But until then, this is the bit where I feel competent and cheerful and brisk. I know where I've been, and I know where I'm going, and hey, it involves books, and we like books.
This is the bit where I don't have to remind myself that this is more fun than any other job I can think of.