I think the whole publishing middlemen part was discussed, at length, by Charles Stross on his blog and shined a light on what's essentially a hidden part of the industry. I can't find it but some Google-Fu might get you there.<p>What I disagree with is where he says: "almost all the profits go directly to the author. I'm not optimistic this will happen any time soon.", which is interesting. Technically there's a few authors who've been self-made, self-published successes through the Kindle store. You lose a percentage to Amazon so can we class them as a publisher? You could chose to sell an ebook directly to your audience as a few people have tried (and have been successful, there's been a few posted on here I've ended up buying very happily).<p>Publishers at the moment are a necessary evil. Traditionally they publish your work, handle the distribution and all the stuff that's basically not the writer sitting their ass down and writing. Digitally they seem to have filled a similar role of providing a digital shelf space, dealing with payments, hosting and so on. Until it becomes completely trivially to write it, host it, charge it and deliver it they'll be needed.
Jeff's points ring close to home. Last year I wrote O'Reilly's Redis Cookbook. Even though I love the guys at O'Reilly (I really do, these folks are amazing at what they do), writing a book provides little direct return. It's a great experience, and I wouldn't say I regret it, but it is far less glamorous than what it might seem at a first glance.<p>There's a comment in this thread that talks about publishers as VCs - which is a correct assessment. Some books do well, some don't. These days, technical books are often part of the latter group, as the amount of information out there renders the vast majority of these books useless.<p>Would I write another book? Sure. In fact, I'm writing the second edition of Redis Cookbook now. However, my expectations are not high. I don't expect to make a ton of money with it. I just want the people who buy the book to be happy with the content, and to contribute to the documentation on Redis. If I make a bit of cash doing so, perfect.
> Lower your expectations. The happiest authors are the ones that don't expect much.<p>I think that's good advice.<p>I'm writing my first book, right now, at 24, on HTML5. (any advice or other articles like this by the way are vastly appreciated) My advance is larger than Resig's (!? what?) but I'm not doing it for the money. Money merely justifies a deadline, not the end product. The end product, to me, is justified by two things:<p>1. Seeing my name "in the wild", in a bookstore, would make me pleased as punch<p>2. If just one person, somewhere, at some point, sends me an email saying they enjoyed the book then that's enough. That's the only encouragement I need to help others (for free) on StackOverflow [1], and its the only thing I need here.<p>[1] <a href="http://i.imgur.com/POZmt.png" rel="nofollow">http://i.imgur.com/POZmt.png</a>
The problem with the move to ebooks for technical subjects will be the lack of editorial oversight.<p>Tutorials, blog posts and a smattering of example code is fine for messing about with some new technology. But where are the authoritative sources. Looking at my bookshelf now, where are the websites that contain all the information I got from (the excellent) "Mastering Regular Expressions" or "CSS: The Definitive Guide" which sits nicely between the CSS spec and actually being readable.<p>These books still need to be written, in the same format, by the same experts because they contain a lot of value. I agree though that they don't need to be physical books -- all my new O'Reilly books I read in PDF format.<p>(Realise there's a lot of O'Reilly naming here, it's only because I really like the quality of their books. I'm not affiliated with them.)
I contributed a few chapters to a technical book about 10 years ago and hadn't really thought about whether I regretted having done it until I read this post. It was definitely a serious pain in the butt but how many of those trials shape who you are?<p>Pros:<p><pre><code> - being abroad in a bookshop and finding 'my' book on the shelf
- sneaking in a picture of my cat (colour scheme example...)
- writing a silly bio
</code></pre>
Cons:<p><pre><code> - the deadlines
- near mental breakdown; felt like my brain was melting
- after the initial writing, the edits arrive
- shoddy advance
- no hope of royalties with a book that was quickly dated</code></pre>
A couple of years ago, I was solicited to write for a traditional publisher. I contributed a chapter to one book and got partway through writing another before I realized that everything related to traditional publishers was awful and jumped ship. I spent a good part of the next year struggling to get paid for my work according to the contract we had signed, and it <i>still</i> pops up and irritates me from time to time to this day. (For example, the time when they accidentally overstated my royalty income to the IRS by a factor of 4 was really entertaining.)<p>Later on, I self-published a collection of my blog posts. The experience was the complete opposite. It was fun, the finished product was much better (technical publishers, oddly, appear to have approximately no idea how to create good technical content), and people seem to appreciate it more. I made as much money, if not more, than I would have selling through a traditional publisher. Although volume is considerably lower, my share of the proceeds is <i>vastly</i> higher.<p>I plan to begin a second self-published book (another collection of more recent blog posts) soon. I will only consider working with a traditional publisher again if the only alternative is begging on the streets, and even then I'll have to take a long while to think about which to choose.
This is exactly the problem we're trying to solve at Leanpub.<p>1) We pay 90% royalty, not 10% (with a flat fee of $0.50 per purchase)<p>2) We think you should start selling your book before it's done, so that you can find out if you have a market for it, and so that you get feedback from your readers as you are writing.<p>3) Once someone buys your book on Leanpub, they get updates to the book for free. This is great for technical books, as you can keep your book up to date when technology changes.<p>4) You own your book. You are free to get a publishing deal with a "real" publisher once the book is done (if you can stomach the hit to your royalties). Imagine going to a publisher and being able to say: the book is done. I have sold 2000 copies on Leanpub, and have tons of great feedback. Here is the manuscript. Oh, and right now I make 90% on every sale. I bet that gets you a better royalty rate than approaching a publisher with just an idea. We all know how much those are worth :).<p>5) We think our workflow is the least painful way to write a book right now, and it's getting better every day. You write in Markdown in a text editor, save to a Dropbox folder, and hit the publish button on Leanpub.
Publishers are a bit like VCs. They invest in a number of different books hoping that one will become a bestseller. All the other ones are either complete failures or midlisters (books which are not bestsellers but are strong enough to economically justify their publication).<p>If you have a bestseller under your belt I guess a publisher makes sense because you would presumably have earned enough money to let other do the boring jobs for you. On the other hand, if you don't break out of the midlist, it may make sense to remove the middle man and go for it alone if you want to write for a living. All the services provided by publishers can be contracted out, specially if you go the ebook way.<p>Finally, I am not a writer but I helped publish several books. Here is a link to someone with a lot of experience and numbers to back it up:<p><a href="http://jakonrath.blogspot.co.uk/" rel="nofollow">http://jakonrath.blogspot.co.uk/</a>
Surprised no one's mentioned Zed Shaw or his Learn Python the Hard Way series...<p><a href="http://learncodethehardway.org/" rel="nofollow">http://learncodethehardway.org/</a><p>According to his homepage, his books have been downloaded 500,000 times. Don't know if those are all paying charges but he also sells online tutorials.
Shouldn't books have github/bitbucket/etc repositories? With readme documents, etc? Hook ups to PivotalTracker with commits published to the github page or marketing site, or pushed to Twitter via IFTTT?