Log in

bear by san

January 2017



Powered by LiveJournal.com
twain & tesla

a letter in your writing doesn't mean you're not dead

I delivered the publication version of Steles of the Sky at about 4 AM this morning. It goes to production now, and if all goes well, next April you will have a lovely bundle of pages or pixels in your hot little hand. (I've already seen sketches for the cover art. Oh man. This is not going to suck. The roughs are currently my computer desktop. Thank you, Donato Giancola, for another gorgeous cover.)

It's the longest of the three books at 153,000 words Microsoft word count, and about 190,000 words manuscript word count.

Irene Cornyn  (LOC)

Why the big disparity? Well, here. Because I keep getting asked what the difference between a word processor's word count and "manuscript word count" is, I explain.

No, take too long. I sum up.

Your word processor will generally give you a word count when you click the proper button. This word count will vary from word processor to word processor, because Scrivener uses a different algorithm than MS Word than Wordperfect than Open Office than what have you. How can algorithms vary? Because wordcount is a judgment call. Is fire-fighter one word or two? Do we count letters and punctuation? MS Word thinks a hash mark for a scene break is a word. Not all word processors agree.

Manuscript word count is calculated a different way. It is not a measure of the number of words in a piece, but the number of "words," where "words" are bundles of five characters and a space. Like column inches, it's meant to tell an editor or a compositor how much space the piece is going to take up in a finished work. So a page of dialogue may have far fewer words than a page of description, but the same number of "words." (Dialogue, you see, has a lot of white space.)

In the era of typewriters, you didn't calculate your word count by counting every word on the page. You calculated it by setting your margins such that your page had either 24 60 character lines on it, or 25 50 character lines. This gives you (very roughly), ~250 words to a page. (This is one reason why standard manuscript formatting calls for a fixed width serif font such as Courier, double-spaced. Another reason is that fixed width serif fonts, double-spaced, give editors and copyeditors and compositors room to scribble all over your manuscript. Another reason is that you get used to reading it, and picking out errors becomes easier. Proportional fonts (where the letters are different widths) look prettier and save space, but also hide errors and make it harder to guess how much space (how many pages or column inches) a given story will take. Sans-serif fonts make it harder to tell a 1 from an I from an l.

In the ebook/webzine era, the industry seems to be transitioning from manuscript word count to MS word count in general--both in terms of payment for stories that pay by the word (alas! because we get paid more for manuscript counts: they average 15% higher, especially if you're a sesquipedalian bastard like me) and in terms of award eligibility categories.

It doesn't matter so much with novels, where you're not paid by the word and the chief issue is the price of printing additional signatures. (Google it.)

But I find I can't accurately compare word I do now with work I did ten years ago unless I keep track of the manuscript word count.

(By the way, this makes Steles of the Sky the second-longest book I have ever written, after The Stratford Man. Which was a 290,000 word monster eventually published in two volumes: Ink and Steel and Hell and Earth. The manuscript was 1190 pages long.)



you're still l337 to me.
It also is a problem when "clever" editors try to predict the page count of a book. I have a couple at my place of employ who reason that if a previous book had 100K words (by MSWord), and this book has 100K words, the new book should have the same number of pages as the previous book.

Certain editors do not understand when I explain it to them. Considering how authors catch on very quickly when I give them the same explanation, I must conclude that some of the editors I work with don't want to understand.

(This has been my "completely unprofessional whine in someone else's comment thread" for the week.)
Because 100K MS word of Robert B Parker is not the same thing as 100K MS word of Neal Stephenson. That's why. *g*
LOL. I used to have a party trick, where I could pick up a manuscript that was printed in 12 pt Courier, look at a couple pages, and tell you how many typeset pages it would be. I was pretty accurate.

I credit it to having done actual old-fashioned cast-offs in my early days as an editorial assistant. But times change, and it just doesn't work anymore.

How do you calculate the manuscript word count? By pages after setting up the margins? (I am groggy so I may have missed it)
If it makes you feel any better, I also still rely on ms. word count as my accurate barometer (or in most cases, my accurate way of charging customers) for copyediting. Microsoft Word's word count drives me nuts, as the algorithm has varied from version to version, and most houses don't recognize or even know that.
Congratulations! You've also just explained something that has long bothered me: how did editors work out how to pay their authors in the old days when they paid by the word.
Huzzah, sesquipedalian!


Also, bravo for finishing Xeno's Manuscript, and thanks for explaining one of the less obvious bits of the publishing industry.