iflowfor8hours/gist:79857f398aa417eee243

## gistfile1.txt
A recent thread asked about book writing and as part of my response to that I decided to link to some general advice about book writing that I wrote down a while ago. However I couldn't find it on jive (maybe it was on wazokazi?) so I thought I'd copy it here for future reference.

I’ve had a few conversations recently with people interested in writing a book and looking to find a publisher. So I thought I should write some of this down.

There are lots of different ways this can happen, and there are variations between due to different kinds of books and different publishers. So don’t take this as absolute advice, just my experience.

First off, remember that writing a book is hard. Everyone I know who’s written one were surprised by how hard it was, even though everyone they knew told them it would be hard. It’s a lot of work and very consuming. In addition it doesn’t pay well. Although you can be lucky and write a best seller (as has happened to me) it’s a rare event. At a rough guess you’ll get something like 5-10% of the cover price of the book in royalties, say $2-4. I’ve heard that a book breaks even at around 5000 copies over its life. Less than half do that. It’s considered a success if it does 10,000, and it’s a best seller if it does 20,000. So if you do the math you’ll see why you shouldn’t write a book for the money. If the book is successful it will boost your reputation, but even that’s not a certainty. So the main reason to do a book is that you care about writing it, it’s something you’re interested in doing and passionate about communicating with others.

Books can either be initiated by the author or by the publisher. In the latter case a publisher says “I’ve got this idea for a book, would you like to write it?” This is not uncommon, that’s how UML Distilled happened to me, and considering how successful it’s been, I can’t knock it. Usually this kind of book is, like UML Distilled, a manual to some form of technology. It’s usually time sensitive, because they want to get the book out before others do.

With a publisher-initiated book the big question is whether you want to do it. You have to be really interested in the subject matter to get through this and usually they’ll need you to work quickly.

With author initiated books, you come up with the idea and look for a publisher to publish it. Again the key is to have a subject that you care a lot about. I also encourage people to pick a topic that isn’t already well covered. You really want something with no competition.

If a publisher suggests a book idea to you, there’s no reason to follow through with that publisher. If it’s a topic that really interests you, you can take the idea and follow it up as an author initiated idea.

The first question about approaching publishers is when. There are two routes here, either early before you’re done much work on the book, or late when you’re close to having a complete first draft. The risk with an early approach is that you won’t actually get the book done, which actually is quite a common occurrence. if you go late you run the risk of publishers not being interested, although in our field I suspect that’s pretty rare. I’ve always approached publishers late, when I’m close to that first draft. That way I know what I have and what the book will look like, as well as knowing that the book will get done.

Next is who, which publishers should you approach? Here you want to look at what publishers publish certain kinds of book. So my kind of technology-independent design books tend to be done by Addison-Wesley, Prentice Hall (who are actually the same as AW), Wiley, O’Reilly, or the Pragmatic Programmers. A book on a java API might be done by these, but also Manning is active in that area. These groupings do change, O’Reilly only got into the ‘software engineering’ market (meaning design and process books) in the last couple of years. Look for publishers that have published some of the books that you respect.

Certainly talk to me, I have a fair few contacts and will happily give you the benefit of my experiences. We have other published authors at TW too - don’t be shy about asking for tips.

Who is actually a bit more complex than that. You often don’t so much work with a publisher as an Acquisition Editor - the particular person who handles you one you work with a publisher. Often the difference between one editor and another can be critical, even at the same publisher. Again asking me for particular tips is one way to find who is worth working with. Sometimes they mention the editor’s name on a book, but not always.

I always advise you to approach multiple publishers. That way you’ll have a choice, both at the contract level and in getting a sense of what the editor is like to work with. Ask them about the process, how do they like to do reviews etc? I take the view that you can’t get enough technical review, so I like at least two rounds with half a dozen to a dozen reviewers on each round.

Generally what you send to publishers is a proposal. This gives an outline of the book, the likely audience, competitive books, etc. Often publishers have their own proposal outlines, you don’t need to stick totally to them, but do try to answer the questions that the proposal outline implies. You can also get a book on how to write book proposals. “Nonfiction book proposals anybody can write” by Elizabeth Lyon (http://www.amazon.com/exec/obidos/tg/detail/-/039952827X) is a good one. You should also provide some sample chapters in the proposal. That way the reviewers of the proposal can get a sense of your writing style.

Another route these days is to self-publish. Leanpub seems a popular way to go. For ebooks you can often produce your own epub and mobi files yourself. The nice thing about this is that you can sell ebooks at a lower cost than publishers will, which may reach more readers. You may also do better financially since publishers do take a large cut. But the bottleneck with readership isn't the cover price, it's the time it takes to read the book. I haven't tried self-publishing (although I might for a future project), some ThoughtWorkers have and seem happy with it.

That’ll do for the moment. If you are interested in writing, speaking etc and have questions don’t hesitate to talk to me. An important part of what I do in ThoughtWorks is to try and help people become more influential in the software community through speaking and writing.
	A recent thread asked about book writing and as part of my response to that I decided to link to some general advice about book writing that I wrote down a while ago. However I couldn't find it on jive (maybe it was on wazokazi?) so I thought I'd copy it here for future reference.

	I’ve had a few conversations recently with people interested in writing a book and looking to find a publisher. So I thought I should write some of this down.

	There are lots of different ways this can happen, and there are variations between due to different kinds of books and different publishers. So don’t take this as absolute advice, just my experience.

	First off, remember that writing a book is hard. Everyone I know who’s written one were surprised by how hard it was, even though everyone they knew told them it would be hard. It’s a lot of work and very consuming. In addition it doesn’t pay well. Although you can be lucky and write a best seller (as has happened to me) it’s a rare event. At a rough guess you’ll get something like 5-10% of the cover price of the book in royalties, say $2-4. I’ve heard that a book breaks even at around 5000 copies over its life. Less than half do that. It’s considered a success if it does 10,000, and it’s a best seller if it does 20,000. So if you do the math you’ll see why you shouldn’t write a book for the money. If the book is successful it will boost your reputation, but even that’s not a certainty. So the main reason to do a book is that you care about writing it, it’s something you’re interested in doing and passionate about communicating with others.

	Books can either be initiated by the author or by the publisher. In the latter case a publisher says “I’ve got this idea for a book, would you like to write it?” This is not uncommon, that’s how UML Distilled happened to me, and considering how successful it’s been, I can’t knock it. Usually this kind of book is, like UML Distilled, a manual to some form of technology. It’s usually time sensitive, because they want to get the book out before others do.

	With a publisher-initiated book the big question is whether you want to do it. You have to be really interested in the subject matter to get through this and usually they’ll need you to work quickly.

	With author initiated books, you come up with the idea and look for a publisher to publish it. Again the key is to have a subject that you care a lot about. I also encourage people to pick a topic that isn’t already well covered. You really want something with no competition.

	If a publisher suggests a book idea to you, there’s no reason to follow through with that publisher. If it’s a topic that really interests you, you can take the idea and follow it up as an author initiated idea.

	The first question about approaching publishers is when. There are two routes here, either early before you’re done much work on the book, or late when you’re close to having a complete first draft. The risk with an early approach is that you won’t actually get the book done, which actually is quite a common occurrence. if you go late you run the risk of publishers not being interested, although in our field I suspect that’s pretty rare. I’ve always approached publishers late, when I’m close to that first draft. That way I know what I have and what the book will look like, as well as knowing that the book will get done.

	Next is who, which publishers should you approach? Here you want to look at what publishers publish certain kinds of book. So my kind of technology-independent design books tend to be done by Addison-Wesley, Prentice Hall (who are actually the same as AW), Wiley, O’Reilly, or the Pragmatic Programmers. A book on a java API might be done by these, but also Manning is active in that area. These groupings do change, O’Reilly only got into the ‘software engineering’ market (meaning design and process books) in the last couple of years. Look for publishers that have published some of the books that you respect.

	Certainly talk to me, I have a fair few contacts and will happily give you the benefit of my experiences. We have other published authors at TW too - don’t be shy about asking for tips.

	Who is actually a bit more complex than that. You often don’t so much work with a publisher as an Acquisition Editor - the particular person who handles you one you work with a publisher. Often the difference between one editor and another can be critical, even at the same publisher. Again asking me for particular tips is one way to find who is worth working with. Sometimes they mention the editor’s name on a book, but not always.

	I always advise you to approach multiple publishers. That way you’ll have a choice, both at the contract level and in getting a sense of what the editor is like to work with. Ask them about the process, how do they like to do reviews etc? I take the view that you can’t get enough technical review, so I like at least two rounds with half a dozen to a dozen reviewers on each round.

	Generally what you send to publishers is a proposal. This gives an outline of the book, the likely audience, competitive books, etc. Often publishers have their own proposal outlines, you don’t need to stick totally to them, but do try to answer the questions that the proposal outline implies. You can also get a book on how to write book proposals. “Nonfiction book proposals anybody can write” by Elizabeth Lyon (http://www.amazon.com/exec/obidos/tg/detail/-/039952827X) is a good one. You should also provide some sample chapters in the proposal. That way the reviewers of the proposal can get a sense of your writing style.

	Another route these days is to self-publish. Leanpub seems a popular way to go. For ebooks you can often produce your own epub and mobi files yourself. The nice thing about this is that you can sell ebooks at a lower cost than publishers will, which may reach more readers. You may also do better financially since publishers do take a large cut. But the bottleneck with readership isn't the cover price, it's the time it takes to read the book. I haven't tried self-publishing (although I might for a future project), some ThoughtWorkers have and seem happy with it.

	That’ll do for the moment. If you are interested in writing, speaking etc and have questions don’t hesitate to talk to me. An important part of what I do in ThoughtWorks is to try and help people become more influential in the software community through speaking and writing.