. . . and another thing I found out about my book

[ritaxis]

In order to find page 77, I had to make a master diocument, right?  so now I know the total wordage of the book, and I'm really discouraged.  It's at 93K and I wanted it to be between 90-120K total and I haven't even gotten to the "midpoint" which I believe ought to be physically at least before two-thirds of the book.  I am almost there, but even so, that would mean the book will have to be at least 150K which is longer than I want.  I wanted a short book!

And my efforts at cutting the early parts were fruitless before.

On the other hand, I do now have a master document, which means when I am procrastinating I can check for consistency in spelling of names, and so on.

Comments

[pameladean]

I'm glad I'm not the only one who procrastinates about making the master document.

And I certainly sympathize with the desire for a short book and the difficulty of writing one.

P.

[thomasyan.livejournal.com]

How do you do that consistency check?

What I've done is conceptually similar to: Convert to a text document, change all spaces into carriage returns (to put each word on its own line), sort each line and remove duplicates. Then I manually scan that list to look for oopsies.

[ritaxis.livejournal.com]

Well, I tend to take the part of the name that I am not so dubious about and use the find function to look for all the instances of the name and look at each one to see if the dubious part is the same. Like there's an ethnicity that I think I wrote Pansky in some places and Pansi in other places. So I'll search for pansk.

I don't understand your method at all. It scares me from the second step: 100K words each on their own line?

[thomasyan.livejournal.com]

(I'm glad you didn't lose all your work!)

With my method, I definitely would be careful what tools I use. I would not use Microsoft Word to put 100K words each on their own line. But there are fairly simple command-line tools to do search and replace, sort, and remove duplicates. And these tools do not modify the original document, so I would not worry about accidentally corrupting or deleting the original.

Applying my method to "How do yo do that consistency check?", where also most punctuation is converted to newlines, too, I would first get:

How
do
you
do
that
consistency
check

and then sorting and removing duplicates would yield

check
consistency
do
How
that
you

where the two "do"'s are consolidated into just one copy. (The tool to remove duplicates could also give frequency counts, which is sometimes nice to have. That way, if 3/5 the time you have "supercede" and the other 2/5 "supersede", that can weigh into your decision of which one to pick.)