Let’s suppose you have an idea for a book, and we like it. We’re talking seriously about it. What do you have to look forward to? On this page I’ll describe the stages books go through and everything you have to do to make the book a reality.
It’s a big job, creating a book. I would venture to say it always takes longer than the author expects. I’m saying this up front because I want to avoid the situation where an author signs up to do a book and then finds he or she doesn’t have time or just can’t work up the motivation week after week.
Before you propose a book to us, try to estimate the number of pages it will be, and how long it will take you to write it based on work with papers or other writing you have done. In today’s fast-moving computer field, a book should be ready to go to production in a few months—a very demanding schedule. We can help you narrow the scope, if necessary, to get the book out quickly. But always be realistic about the amount of time you can spend.
In particular, great persistence is a part of doing a meaningful book. We avoid the easy-to-write books one finds all too often on the shelves; we don’t want books that say the same things one could find in vendor or online documentation, or that breezily describe an activity as if it always works the first time, hiding the thorns that all-too-often prick the unwarned user.
One of my authors built a number of systems over and over in every manner he could find, just to understand all the twists and turns of building such a system. He also faithfully followed the mailing list devoted to those issues for months. Even the experts on the mailing list often admitted they didn’t really understand the process and considered it a matter of luck when it works. Finally he felt he untangled all the kinks and could write the details people need. Meanwhile, an author writing a competing book chose to document just one commercial product, and pretended to cover the build process by describing a script that came with that product. A half-hour’s work versus several weeks’ worth. Well, at O’Reilly we want you to do the several weeks’ worth of work.
You should also build in some extra time for responding to my requests as your editor. As I explain in my page on “How we regard our work,” O’Reilly editors get heavily involved in order to ensure the quality and feel of our books. It’s going to take some time for you and me to agree on how to introduce ideas and organize the book. The first couple chapters, at least, will probably go back and forth between us twice; the first time I may ask for a complete rewrite. Build this adjustment time into your schedule. And remember that I’ll always find extra topics to ask for and stylistic changes to make, even after I like your general approach. So add at least a few days per chapter to your schedule, to accommodate rewrites.
When you’re ready to make a proposal, send whatever ideas you have to proposals@oreilly.com. (You can also contact me or another editor directly, if you think we’d be particularly interested.) Your proposal may be just a few sentences long, or you can include an outline (we’ll ask for one eventually) and even writing samples.
It’s probably not worthwhile coming to us with a book you’ve already completed. Because the O’Reilly approach springs from a lot of subtle attention to detail, we like to start from scratch. We hardly ever (I’d almost say never) accept a manuscript that was written without strong input from one of our editors.
Our proposal process is informal, and we always try to have an open dialog. The editors discuss proposals among ourselves, and sometimes bring in opinions from experts we know inside and outside the company. We may transfer your proposal to another editor whose background is more suitable. If we think you have some valuable insights but don’t think the exact topic you’re proposing will sell books, we may talk to you for a long time to find just the right topic. Eventually we’ll settle down on a contract, but this may take several weeks.
When you and the editor agree on what to do, the information we need to draw up a contract is:
We won’t hold you strictly to any of these. They just make sure that you and we agree that you’re writing what you said you’d write. Ultimately, the question is whether the editor thinks you’re producing useful and readable material.
We don’t always give advances because we expect projects to be finished and released soon. Sometimes we can pay a few thousand dollars over the course of the project. They can help you buy a piece of equipment or software you need to write the book, but they won’t allow you to take a lot of time off from work. You have to find extra time to write.
Forcing yourself to sit down and write, week after week, is the hardest part of the authoring process. Popular definitions of genius are relevant here (like “the application of the seat of the pants to the back of the chair” and Thomas Edison’s: “One percent inspiration and 99 percent perspiration”). Many authors get up two hours earlier than usual and find a quiet place to write. You may find time in other ways, but be sure to set some aside every week.
A couple of my authors have started new jobs (at start-up companies, no less!) in the final weeks of writing their books. Not a choice I would recommend. In both cases it caused a lot of stress as they rushed to get their final drafts to production.
We try to let authors write with any tool they like. But we have developed a standards-based platform that most authors find easy to use and that is tailored toward producing O’Reilly books: Atlas. It stores your text internally in an enhanced HTML format (which you can view and edit, if you want) and can produce all the formats we distribute.
What’s really important is not the tool you use, but the book’s format. We have designed a special look for our books (headings, fonts, and so forth). These are not things that most authors worry about, and our design is widely seen as attractive and effective. What you have to do stick to the elements of our design. Everything you should need is there, including classy stuff like sidebars. Whatever format you and we settle on, we’ll provide you with templates or macros that automatically keep you within the right design. You can even print or view a document exactly the way it will look when we publish it.
As you send me chapters and we discuss them, we are building up a rapport. I view the relationship between author and editor as one of mutual enlightenment, where we gradually come together in our understanding of what to do with a book. You are educating me about the technology and what its users care about, while I am educating you about how to draw readers in, make them want to read further, and organize information effectively.
Your first couple chapters may have to be rewritten substantially, before we reach a common viewpoint. But many authors find this process exciting, and it’s wonderful to see the way things perk along when we both know what we’re doing.
I sometimes do two reviews of each chapter (at least near the beginning). The more comments I make on the draft (or online document) the more I like it. The first time around, I may ask for a few large changes, like how the chapter is organized, what information you cover, and how you approach your readers. I may say, “You have to discuss M before you mention N” or “Leave this for another chapter.”
After I’m satisfied with the overall flow of the chapter, I zoom in on the details of each paragraph. I do a lot of copy-editing at this point: points of style, grammar, and so on. You’ll get another pass at this from a copy-editor when the final draft goes to our production group. But I think it helps the technical review to have as readable a draft as possible; mistakes and awkward phrasing distract reviewers.
We give writers an unusual amount of support at O’Reilly. One author I’m working with had absolutely terrible writing (I don’t know how he managed to produce a readable proposal, but he did). His spelling, grammar, wording, and everything else about style was riddled with errors. But I didn’t mind correcting all this (as tough as it was) because his thinking was crystalline. He knew absolutely what to tell users at each point.
Besides your text, don’t forget some other important parts of your document:
We now create most indexes here (and we don’t charge you for doing so, like some publishers). You’re welcome to put in index hits if you think you can make subtle connections that a non-expert wouldn’t find.
Authors usually, at some point, get bogged down and have trouble writing. It’s usually not at the beginning; more likely it’s about half-way through, when they’ve lost the initial excitement that gave the project momentum and when the end of the project is too far off to provide the bit of second-wind energy that comes from knowing you’ve almost reached your goal.
I raise this warning now so we can both recognize it if it happens to you. You have to start your project knowing that it’s a major commitment and that you may sometimes rue your impetuosity in taking it on. And yet it is worth the effort, so long as you push through the period when you don’t really feel like writing. A few chapters before the end, nearly every author picks up speed and feels like he or she can focus on exactly what to say.
Another phenomenon that generally marks a milestone is the moment when the book goes public. Publishers have wonderful opportunities to get special promotions by booksellers and other publicity. So at regularly intervals I’m asked which books to announce; books I can commit to producing. This is usually well before release dates, before the books are ready for technical review, so I have a frank talk with my authors and decide whether they are certain they can meet their deadlines and feel ready to announce the book. If we decide to announce it, we can’t use too many excuses for being late. Up to that point, the authors owe nothing to anybody but me; now they owe their book to our marketing and sales team, to the booksellers we’ve announced it to, and to the masses (we hope) of readers waiting for the valuable material.
Here, “first draft” means a document that is ready for reviewers to see—it’s not the first impressions off the top of your head. As we have seen, you could put your work through a number of revisions before it reaches the “first draft” stage.
Now you can take a breather. The book goes to technical reviewers, some chosen by you and some by me. The course at this point is:
So it’s three or four months between the time you finish the first draft and the time it goes out to the public. Most of this period doesn’t require much input from you. During the technical review period, you can be researching the topic for last-minute updates.
We usually get three to five technical reviewers. Some of them will never return any comments at all—that’s a fact of life. It’s OK; maybe they’ll at least tell all their colleagues that a great book is about to come out. If a couple reviewers do a thorough job, we can be confident that your book will satisfy its audience.
Usually, incorporating reviewer comments is straightforward and requires only minor research. Once in a while, a reviewer points out the need for major changes. Perhaps you didn’t think about some segment of your audience. Or maybe there’s another way to do things that you didn’t know about.
This happened to me when I was an author; a single reviewer of Managing Projects with make pointed out all sorts of things that were different on versions of make I had never used. We spent the next month exchanging shell scripts and trying out features on different versions. I stayed up a lot of late nights, but I got the job done and I am very grateful to that dedicated reviewer.
If you have something like this happen, your month of changes is going to be a busy one. We may even have to slip the schedule, although we hate to do that because it throws a monkey wrench in the marketing effort, which will already have begun.
When your final draft is done, we turn it over to production. There it undergoes a copy edit, which often turns up interesting problems. The copy editor makes sure that the style meets professional standards, and takes care of all the issues of consistency that authors (and I as well) tend to forget about in the heat of the creative act. Usually I can handle questions that copy editors have. Occasionally we have to go to the author to resolve some ambiguity or supply missing information. Most of the time, though, you’re out of the hot seat.
If your figures weren’t finalized before, that will happen now. We’ll show them to you for review. And we’ll work some more on your index.
Since you feel an intense connection to the manuscript on which you’ve worked so hard, it’s natural to worry about what’s going on in production. Some authors just ask why this stage has to take so long, while some fear that we’ll introduce errors, and some ask for unusual stylistic features.
Actually, we have an unusually fast turn-around for the book publishing industry; your book doesn’t sit around for months as it does in many other places. But we never skimp on quality checks.
We don’t encourage you to add features to our design and format, but if there’s a really good reason to do so, I’ll put you in touch with the design and production people who can implement it. But you’ll have to accept that some things are too costly. The cost may not be apparent to you, but books are like software in this regard—the developer often doesn’t understand the whole maintenance cycle.
All during this time, we’ll be marketing the book. We’re careful not to market a book too early, but we want the public to be prepared for it when it hits the stands.
And as for you, once you get over the feeling of burn-out, you can start to think of doing another book!