Latest from mblog Random from mblog


Ta-da List

RSS 1.0 / 0.91

March 24 2006, 21:26

Shaking up tech publishing

I've been talking to a lot of friends and acquaintances who are writing tech books for a wide variety of old-school publishers and I can't believe the deals they're taking.

It seems that the industry standard is something akin to 10% of the profits (which easily take 4-5-6 months to arrive), being forced to write in Word, and finally a production cycle that's at least a good 3 months from final book to delivery. That's horrible!

And what do you get in return? Usually not all that much. There's rarely a big marketing push to be had and you're expected to do lots of the editing yourself. So you get some editing, a cover/layout, and the distribution done for you. Is that worth 90% of the profits and the torture of writing a book in Word and then bouncing versioned documents back and forth?

The standard sugar coating of this setup is that you should not expect to make money writing a tech book. That it's not about the money, but the fame and authority and satisfaction of seeing your name in print. While all of those things certainly do have value, why on earth would you want to accept the premise that writing a book is not going to be worth it for the money?!

Of course it's not going to be for the money when you only land 1/10th of the crumbs that trickle back to the publishers from Amazon and independent retailers. Especially in tech land where its very rare to make it up on volume (your book is not going to be in every airport around the world).

No, we have to change the game such that each copy becomes much more profitable. Just accept the fact that you're not going to sell 100,000 copies and optimize for the scenario that you might sell 5,000 or 10,000 copies. It is entirely possible to make money at those levels, but not if you have a traditional cost structure.

Self-publishing is easier than ever
At 37signals, we optimized for a niche seller by writing all of the content for Getting Real in Writeboard. Then Jason spent a little week marking those 200 pages up in InDesign. Then we spent a day building a dead-simple shop in Rails that would take $19 from your credit card and give you a PDF.

That scenario worked well for Getting Real, which was all short essays that you could print out if you liked or even read on the screen without too much distress. And in just three weeks since its release, we've sold over 5,000 copies. You do the math. Even at such "low" levels, it's comfortably profitable. It's worth doing it for the money.

Lean publishers can make deals not suck
But you don't have to abandon dead trees entirely to be profitable as a tech author. Look at the Pragmatic Programmers. They use Subversion for revision control and collaboration alongside a tech-powered writing pipeline where authors write book "source code" and are able to produce PDFs straight from that! There is not even an InDesign step at the end. And you get to write in your favorite editor (I used TextMate for my work on the Rails book).

With drastically lowered cost of production, they also get to offer authors a much better deal than the industry standard. They simply share the profits with you. Fifty-fifty. And since they sell PDFs and paper-versions direct, the pie is much larger than when you only go through book stores and Amazon, which heavily discount and need their slice too.

And it's not exactly little league numbers they're capable of doing. Agile Web Development with Rails has had combined sales of around 40,000 copies (counting both PDF, paper, direct, and not) in the 8-9 months it has been out. With a deal like the one the pragmatics are offering, that's very real money.

Is the quality suffering from that? I'd say on the contrary. The book just won a freaking Jolt award. And I've received so many emails from people loving it.

Easier, more profitable, and better: Pick three.
So it seems that you can have your cake and eat it too. That's how I feel the proposition of going it on your own or hooking up with a lean publisher looks like.

Disclaimer: No, I don't hate traditional publishers. I know many and they're very nice. I just don't like the deals they forced to offer people. And yes, I did co-write a book for the prags, so I'm biased to like them.

Challenge by Jacques Marneweck on March 24, 22:27

Busy reading the Rails book and have read the Getting Real book from 37 Signals. It's great getting to hear about these stories where authors are moving away from more traditional publishing to using PDF's (which is different as I got my first pdf book experience with php|architect about a year ago).

Challenge by ramanan on March 24, 22:36

The getting real book was the first PDF book I've bought. (I enjoyed it.) I think the fact it was unencumbered by DRM and nonsense like that was a nice touch, and probably helped with your sales. I think in the tech field publishing online (electronically) is quite viable. Your audience is going to be tech-savvy, and probably more inclined to have a positive opinion on electronic books.

Challenge by anil on March 24, 22:51

I would never even consider writing a book in something other than Tex with SVN control. The productivity pluses are so great that i'm even trying to get non-programmer friends (humanities post-grads who write essays/books, journalists etc.) to pick up this way of working. If someone could package these technologies for freelance writers/editors through an online service, i think they'd be onto something. A small but significant step in digital publishing perhaps.

Challenge by cuomo on March 24, 23:00


Challenge by Miles on March 24, 23:02

Have a look at They allow people to write and sell paper versions of their book. They produce books one at a time where the content is printed, bound and then shipped to the buyer. Very cool stuff (and no I don't work for Lulu)

Challenge by rps on March 24, 23:21

From a buyer's point of view, the main thing publishers offer is that they filter out the crap. Self-publishing is easy, but if I don't know who the authors are, I'm not going to want to waste my time on something that's probably drivel.

Somebody could probably do very well starting a new "lean" publishing company that would share more money with authors, while essentially providing nothing but a name on the spine to assure readers that the book isn't crap.

Challenge by Ryan on March 24, 23:44

"tech-powered writing pipeline" or "tex-powered writing pipeline"?

Challenge by Joe on March 24, 23:59

A publisher's name on the spine is no guarantee that it's not crap.

Challenge by Rick on March 25, 0:29

It could be worse. They could force you to use OpenOffice.

Challenge by anil on March 25, 0:42

Top down control structures can't guarantee quality (a publishing house with a good reputation can publish poor books). Bottom up structures can. You don't have to worry so much about quality control with open content - collaborative filtering sees to it. Thanks for the tip, looks interesting.

Challenge by Jack Shedd on March 25, 2:25

There isn't a structure in the world which can "guarantee quality".

Challenge by Frank Mash on March 25, 2:32

Very well said. Traditional publishers are definitely threatened as modern publishing becomes more convenient and affordable.

-- Frank

Challenge by Jan on March 25, 2:44

Personally I wonder in what software they make the head first series at o'reillys. Brilliant books IMO. They are very visual though, so I don't think they did it in either word or a basic text editor.

I have two books from the series, and will probably buy other books in that series if they are interesting. The learning experience is just miles ahead of any book I ever read. (and yes, that includes several of the pragprogs books, and the getting real essays). It is just amazing how much you retain after reading a head first book the first time.

So my advice to anyone wanting to writing and publishing books: be more like kathy sierra. :)

Challenge by Tim O'Reilly on March 25, 4:58

The Head First books are produced in InDesign. They are also a good counterpoint to David's arguments. They were the product of a brilliant author team who created the books largely by themselves. In addition to being a creative genius who came up with a brilliant new teaching methodology, Kathy is also a great promoter, with an amazing blog, and hugely successful training seminars at our conferences. But authors like Kathy and Bert are a rare find. Scaling up the series has been extremely hard. We've had to go through dozens of potential authors, with numerous training sessions, to get people who can write in the style. Even training editors who can supervise potential authors has been a slog.

Some series have simple formulas, but really great ones take a lot of work.

Challenge by Tim O'Reilly on March 25, 5:37

In addition to responding to Jan's comment about Head First, I wanted to respond to David's fundamental premises in this posting. There's a lot that David got right, but also a lot that he got wrong. So I thought I'd share some thoughts from someone who's been there, both as a brash newcomer to publishing with all the same criticisms back in 1985, and now, an example of "the establishment."

At bottom, I believe that you're reasoning from insufficient data, David, and generalizing from experiences that are not necessarily reproducible by other authors. While established publishers are often sclerotic and slow, they aren't all stupid or venal. As businesses get larger, they have to manage their economics by the averages, not by the successes. If we could all pick winners every time, we'd all be rich and living in Lake Wobegon.

So, a bit more detail on where I think you got it wrong, David:

First off, to clarify, royalties for most tech books are typically 10-15% of net sales, not of "profits." That's a big confusion. Most book publishers have profit margins in the 5-15% range, and so 10-15% royalties paid to authors are actually equal to a 50/50 split of profits. What the Prags do isn't a 50% royalty either--it's 50% of net revenues after printing and some other expenses. Perhaps a 35-40% royalty -- still darn good -- but not 50%. (For reasons I'll get to later, I don't think that will last, but only time will tell if this is really a new model, or just another example of how the more things change, the more they turn out the same.) The distinction between revenues and profits is very important, and many new businesses aren't clear enough about the difference. In particular, many tasks that are just borne as a matter of course by a founder and are thus "free" become expenses as a business scales and it becomes necessary to hire someone to do those tasks.

How about sales volumes? David's got some false assumptions there too. Very few computer books sell 100,000 copies. Most traditional computer book publishers are in fact already optimized to be profitable on books that sell from 5-10,000 copies. That's the size of the market for most books when the publisher is able to call on all the traditional avenues of distribution. And even then, the odds aren't that good. We get a report from Neilsen Bookscan on the Top 10,000 computer books sold each week, and we load that data each week into a MySQL data mart. We have nearly 20,000 unique books in the database -- computer books published over the last four years. And of those 20,000, only about 300 are currently selling at rates that will give total sales of at least 10,000 copies a year. Granted, some of these books will sell for many years, but others will need to be revised annually. In short, most computer books published sell far fewer than 5-10,000 copies, and publishers are playing the averages. For every 100,000 copy bestseller, there is the 1000 copy failure, with profitability usually found in the 5-10,000 copy range.

Now let's turn to production methods. For many authors, Word is not a requirement of the publisher. It's the preferred method! At O'Reilly, we've struggled with this, and ended up standardizing on Word because it was what most authors want. And to this day, we do in fact work with many different formats, as do many other publishers. (Many publishers in fact take camera-ready copy from their authors, and let them write in whatever tools they like!) So again, I think that David is tilting at the wrong windmill.

As to not much editing...I can't speak for my competitors, but the reason why O'Reilly is so consistent as a publisher is that we do a LOT of editing. Many of the books we publish only become publishable because our editing and production staff put in a lot of time and effort to help an author capture his unique knowledge in an effective and consistent presentation. It's fairly rare to find an author who is both technically competent and able to turn out flawless prose. (I still remember the six years and six authors I spent before finding someone who could do justice to Sendmail.) If most books needed no editing, David would be right that publishers don't add much value. But having published over a thousand books in the past twenty years, I can quite definitively say that that is the exception rather than the rule.

Finally, the long production time. It's true that more agile methods can get books out more quickly, and there's a lot to like about the Prags' methodology...but unless you're only selling direct to the end customer, and not through bookstores, that won't do you any good in reducing time to market. The big chains require not three but four or five months notice on a title before they will stock it, and because of their budgets, they don't take kindly to titles that don't make their projected dates. So cutting production time doesn't really help, unless you can predict when authors will be done with a great deal of accuracy early in the writing process.

And if you just say, "well screw the retailers, I'll just sell direct," you may only be able to sell direct -- not just because they might hold it against you, but because you'll have skimmed the cream off sales, and thus the modeling that retailers do of sell through when a book is first introduced may lead them to purchase far fewer copies than they ought to, thus reducing total sales for the book.

Perhaps, as David says, the alternative is to bypass the entire system, and just go to self-publishing. He uses as an example the success of 37signals' recently published PDF-only book, Getting Real, which he describes as a niche book.

Sorry, David. That isn't a niche seller. That's a bestseller. And why not? 37signals is one of the most celebrated small companies on the net, one whose development practices are widely admired and emulated. You guys are famous! Of course you can sell lots of copies of a book that promises to tell people the secrets of your success.

But this doesn't mean that every author can expect the same results, any more than the next blogger can expect to get the same traffic as BoingBoing or Engadget. Doing everything right is necessary, but not sufficient for success. Some newcomers hit it just right and go to the top of the heap, but most live out their life somewhere in the tail.

OK. How about your next example, your book on Ruby on Rails with the Pragmatic Programmers.

Again, David, you're generalizing from a bestseller, not the normal case. Agile Web Development with Rails is currently #26 on the Bookscan list of all computer books. At its peak sales period, it was in the top ten. Counting its direct sales (not showing in Bookscan), it might well have been in the top three or four.

So, here's how I parse your comments: If you hit the wave just right, with the first book on a really hot new technology, from the creator of that technology, you can have a really successful book. And if you're lucky enough to co-author it with the owner of a small company that offers very good royalty rates, you're really in the money.

David's done really well for himself. I can't disagree. But I am concerned about the overall message.


First, I believe that there will be a lot of disappointment from people who expect similar results. The average self-published book might sell a few hundred copies, a successful one a few thousand. It won't sell thousands in a few weeks. Very few will have the success that 37signals' self-published book has had. Meanwhile, authors who are persuaded that their publishers are screwing them will end up not taking a route that may in fact be better for them.

And as to the expectation that a 50% royalty rate is fair, I'll remind people that APress also started out with a 50% royalty offer, and have over time modified it to a level much more consistent with traditional practice in tech book publishing. Normal royalty levels are not some plot by publishers to screw authors: they are a reflection of the real economics of the business.

Right now, the Pragmatic Programmers are in a unique honeymoon period, with several distortive characteristics. First off, they are themselves the authors or co-authors of six of the twelve books they've published, so a good amount of the royalties they are paying are to themselves. Second, they have had the good fortune to start out with a couple of bestsellers. Third, they haven't been in business long enough to experience what happens when the party ends for a bestselling book.

And it does. It's possible to sell a lot of copies of a book into the retail channel, only to have those come back once sales slow down. This is especially true for bestsellers. A really good example is a book we published back in 1994, Managing Internet Information Services. It was a hot follow-on to our bestselling Whole Internet User's Guide and Catalog, which was selling 250,000 copies a year. MIIS, the administrator's guide to the net, sold through 50,000 copies its first year. But over-excited by the success of the Whole Internet, retailers had bought well over 100,000 copies...and sent 50,000 back in returns, suddenly, just as we'd printed another 25,000 to satisfy what seemed to be continuing demand. And once they decided that the book was a failure (because it had a 50+% return rate), they stopped ordering all together! Markets aren't rational, and eventually the party ends.

But even in less extreme cases, the more successful a book is, the greater the chance that you'll have thousands or even tens of thousands of copies in the channel at the point when sales begin to slow down.

At that point, you get to eat not just all the extra copies you printed, but also perhaps the royalties that you already paid out to the author.

But setting aside these huge problems that come from success, you can't always guess right. A publisher bets on a topic that seems important, produces a great book, and it sells only a few thousand copies. Overall, the royalty rate has to reflect the average expectations, not what would be acceptable if the publisher had all bestsellers.

For many authors, a royalty advance of $8-10,000, plus royalty upside of another $10,000 is more than they'll see from a self-published book that sells 500 or 1000 copies at $20 or $30, after they deduct their manufacturing cost. And if a book really hits, the access to channels can lead to huge upside. I have quite a few authors to whom I've paid hundreds of thousands or even millions of dollars in royalties over the lifetimes of their books.

I believe that the Prags are assuming that every book they publish will be as successful as their first few, and that when that inevitably turns out not to be the case, they will find that in order to stay in business, their royalty rates will eventually need to approximate those of other publishers.

But keep on challenging us, David. It's easy to get complacent. And you're absolutely right that there are new options available to authors, and that self-publishing success can be yours if you're good and determined and lucky. I know. I started as a self-published author, and the rules really aren't that different now than they were back in 1985. For everyone who succeeds, there are a lot more who don't.

And depending on your temperament and your skills, there are still a lot more people who'll be successful working with a publisher than going it alone.

Challenge by David Heinemeier Hansson on March 25, 8:52

Great response, Tim. I firmly agree that it'll be hard to out-O'Reilly O'Reilly. That is think that you can get to the same position and volume simply by being better at spotting best-sellers.

I do think, however, that you can get a different risk profile by lowering the cost of production and chasing new demand streams with greater vigilance (I've been trying to off-load ideas on some of the streams to many O'Reillyans). That's still playing a portfolio game, though, which is not my main argument here.

My argument is that you don't need the scale of O'Reilly to make it big. That choosing to play a portfolio game is often a diWORSEify game (see Peter Lynch).

No, 37signals can't scale to spit out 10 books a year on areas outside our domain. But we don't have to. If we can just write and push the angle we now really well, we have much-better-than-average chances of success.

Likewise, the Pragmatic Bookshelf has a unique position in the Ruby community and better-than-market odds at picking winners in that niche.

So my argument is that whenever you believe that you have "inside information". When you know more about a niche than the market in general, you stand a great chance of beating the market.

And no, I don't believe that neither 37signals nor the prags are unique in their knowledge of a niche (defined as a single subject, not size of that subject). Kathy Sierra is showing a killer niche in her teaching style (this time the niche is approach, not subject). There are undoubtedly others.

What all of these examples share is a required humility. Especially for 37s and prags. We can't branch out and write a book about game development and expect above-market returns because we don't have above-market competency in that niche. If we did, we'd be in the same portfolio game that O'Reilly plays so well (and would have to face reality when the first string of bets went sour as your example with APress shows).

So what I'm suggesting is that we should see a rise of more niche specialists that extract the cream from a subject by relying on their inside knowledge to beat the portfolios.

It won't scale, but it doesn't have to. You don't need a huge operation or nearly as many bets if your just playing your own game.

Challenge by Dae San Hwang on March 25, 9:14

In response to Tim O'Reilly's:

I think innovators like David tend to speak for themselves. From their standpoints, the world (or the industry) always looks like a retard. It's just unbelievably inefficient. Things could be so much more effective if everyone was just like David. ;)

Innovators tend not to do well in/with big organizations. Bureaucracy slows them down. It's inevitable. Big organizations get their competitiveness from their members following the same stable rules. (You see, how each individual ant follow such a simple rule to demonstrate a complex yet very effective emergent behaviors.) Innovative individuals get their competitiveness from abandoning old rules in monthly (or even daily) basis.

While Tim, your comment is much more balanced, more thoughtful, and simply more logical, I think, David is on to something else. He wants to provoke others who may be as smart as him but possibly have been less determined (or just unlucky) to come out and to innovate. Many smart people don't do that either because they are already complacent or doing so seems too risky. He seems to think that world would become a more interesting place if every smart people just try to out-innovate each other.

For some reason, already successful, innovative people seem to want to change the world big time. Observing what people like you, David, or Paul Graham are saying and doing, I guess, you guys simply need bigger challenges. ;)

These are just some of my rambling thoughts. Thanks David for another provocation and thanks Tim for sharing the publishing industry's insider knowledge. I always enjoy what you guys have to say!

Challenge by Ross Murker on March 25, 11:28

So, how does someone running a bare-minimum subsidy publishing service get *seen*?
Check out for an example of what subsidy electronic publishing could be.

Challenge by Tom on March 25, 12:06

It's gracious of Tim not to mention it, but it's fairly clear that the Agile Rails book would have been an order of magnitude more useful if it had been published through O'Reilly. Yeah, it's undeniably useful (if only because the freely-available documentation is so poor), but the level of self-indulgence and lack of organisation is enough to half-cripple the poor thing. A really good editor, who was thinking about what the audience would need on a day-to-day basis (rather than what the authors thought was a good idea at the time), could have improved the book immensely.

So why did I buy the Agile Rails book? Because it was the only one available. That doesn't say much for quality, although it certainly shows that agility is an advantage.

Challenge by Tom on March 25, 12:08

Furthermore: why did it win the Jolt award? Because Rails is a good technology currently surrounded by white-hot buzz, and that book is the only convincing "reference" (I use the word loosely) to what Rails can do. If O'Reilly had published a competing title with the same technical content (e.g. with DHH as coauthor again) but with their own editorial standards applied, Agile Rails wouldn't have got a look-in.

Challenge by Dae San Hwang on March 25, 13:20

I disagree with you Tom.

While publishing through O'Reilly might have improved the quality of the Agile Rails book in some ways, as you suggest, it would not have helped Rails community (and the world for that matter) as much as it has done.

Here's why. First, Pragmatic Programmer already had published the Pickaxe Book. Dave Thomas was the single best known figure in English speaking Ruby community. By co-authoring the book with Dave Thomas, I think Agile Rails book have achieved the highest synergy possible within and outside the Ruby community.

Second, I'm not sure if Rails was big enough in late 2004 for even O'Reillly to consider publishing a book on it. I'm not sure if Rails was even on Tim O'Reilly's radar back then. And even if it was I doubt O'Reilly would have published Rails book without first doing a comprehensive Ruby book first. Yes, how quick Rails has succeeded is unprecedented in the history of technology.

Third, we really need to give credit to David and Dave for doing the beta book for Agile Rails book. Do you really think Rails would have catched on so fast if the Agile Rails book came out only in last December? David and Dave has really pioneered in publishing pdf books in the progress of being written. Even O'Reilly is catching on this trend with their Rough Cuts series.

While editing is an important criteria, for books like Agile Rails speed matters far more. They can redo the editing for the later editions.

With all that said, I think I know what bothers you. But I tell you what, it's the dissatisfaction with the status quo that drives the innovation. I respect O'Reilly as much as others. Heck, I would even place Tim O'Reilly on the list of people who has changed the course of my life. But David just isn't completely happy with the status quo of the tech book publishing. He already did it a big favor by serving his own book as the guinea pig for the beta book experiments.

If you appreciate innovations, don't complain. Just enjoy the revolution that's happening around us.

Challenge by Tom on March 25, 13:41

Like I said, they did well to get it out so quickly, and undoubtedly that was to everyone's ultimate benefit (I'm glad I've had Agile Rails for the last few months, instead of having to wait until now for O'Reilly's comparable offering); I just think it's disingenuous for David to suggest that the intelligent, objective editorial control provided by experienced publishers like O'Reilly has little value. If O'Reilly were more proactive about getting the ball rolling on technologies that are just beginning to appear over the horizon, in combination with their Rough Cuts beta-book ripoff, while (somehow!) maintaining their editorial standards, they'd be routinely blowing the Prags out of the water.

Dave Thomas may be good from a "community" angle, but the prose and structure of his books are poor -- if he was paired with an excellent editor we'd be getting the best of both worlds.

Challenge by Dae San Hwang on March 25, 14:37

While I don't entirely disagree with you, Tom, I think you might be missing the big picture. The approach you suggest may not be pragmatic. Many smart people have strong opinions about how their creation should be like. Some of them wouldn't do it if they have to compromise. Kathy Sierra's Headfirst series almost didn't see the light of the day because most technical editors thought it was a joke. While editors/publishers serve as the crap filter, they also sometimes stifle innovations in their industry.

Even Ruby is all about freeing individual programmers from the random burdens Java language committee would place on average programmers. You wouldn't ban the use of blogs so that every single content on the web are edited by some professional news editors, would you. It's just not practical.

As David often says his stuffs tend to be highly opinionated. While I don't agree with all his opinions, I'm not bothered because I rather hear 100 his thoughts with probability of 20% being wrong than hear 10 his thought with probability of 99% being right. I got a brain to decide them for myself.

By the way, O'Reilly already does exceptionally well in spotting emerging technologies. (They even host conference called Emerging Technology Conference!) Rails was too small a niche even a year ago, or even O'Reilly is too slow in spotting innovations that happen so quickly. We really can't complain. The time we are living is almost ideal! ;)

Challenge by Neil Wilson on March 25, 15:04

My question is why are there tens of thousands of copies of anything in any channel at anyone time. That's a hell of a lot of capital that could end up in the shredder.

The PDF book eliminates that risk, and the beta book cycle allows you to judge print run demand a little better.

It's the paper that is the problem. What is stopping print on demand? Lack of technology, capital, or will?

Challenge by Mr Coffee on March 25, 16:39

Neil, the reason why paper still exists and is still succesful is simple: you can go to that nice coffee place around the corner, cuddle up on the couch there with a nice hot latte and your copy of the Rails book.

Books + Latte = Happy Mr Coffee

And to that i'll celebrate with a hot cup of home brew.

Mr C.

Challenge by Tim O'Reilly on March 25, 17:54

Neil -- The ultimate reasons for paper (from an economic perspective rather than a usability perspective) are still for additional channels of distribution. The ultimate challenge for any information product (once a quality product has been produced) is getting noticed, so the more ways you can get noticed the better (as long as the costs of doing so aren't greater than the rewards.) As I wrote years ago, retail distribution is like alveoli in your lungs -- a mechanism to get more surface area for contact. With the web, there are many more such mechanisms, but they too have gotten crowded and expensive. People pay more today to get their information product in front of customers at Google than they ever paid to Barnes & Noble! (For more of my thoughts on publishing, and e-book publishing in particular, see

Challenge by Tim O'Reilly on March 25, 18:08

David -- I really loved your response to my return screed! You're absolutely right that someone who is the master of a niche can indeed make good money by creating paid PDFs for download as part of their product portfolio. But don't overestimate the size of the population that has all the right skills. One of the hardest lessons in growing my business over the years has been that things that seem easy to one person can be very hard to replicate, because they just don't realize how many talents they bring together in one package. Costs are higher because it takes multiple people to do the task that one person with all the right skills can just dash off. A poorly written or badly organized document that doesn't successfully teach what you know will not only not sell (unless people are really desperate for information), it may damage your reputation.

The point is that there are lots of options to consider, and I wouldn't count out the value-add of a quality publisher, both in developing the product and in helping it to find its widest audience.

It is however true that the economics have tilted in favor of of self-publishing for many topics -- or at least in favor of some kind of pdf-only publishing. Before the dotcom bust, 10,000 copies of a printed book was easy to achieve, and 30-40,000 was fairly common. Now, the ceiling is far lower, and the chance of getting a breakout bestseller that justifies the widest distribution are much smaller. If you're weighing selling 500-1000 copies of your download against 5-10,000 copies of a published book, you've got a different equation than 500-1000 versus 25,000. But even with pdf downloads only, a publisher's reputation and critical mass can help.

Challenge by Bryan on March 25, 18:15

Speaking from the point of view of a reader, I really welcome the change towards more tech books being published as PDF's. Tech books have a short useful period where I read them often and then once I master a subject I find they just take up space and then have to be replaced when newer versions come out.

O'Reilly's Safari web based books are nice for browsing but when you really want to zoom in and read a whole book it's PDF export option is very clumsy and the resulting PDF has poor layout that is nowhere near the quality of the printed book.

Once ePaper book readers come out in force, PDF's should be an even better form of book publishing.

Challenge by Tim O'Reilly on March 25, 18:34

Tom, you refer to O'Reilly's "Rough Cuts beta-book ripoff", as if the Prags had invented the idea of the beta book. Just for the record, we shipped JavaScript: The Definitive Guide as a beta book (using that very name) back in the mid-90s. We didn't continue the practice because the print edition of the book (done as a kind of print on demand) ended up creating marketplace confusion for retailers and lots of returns when the final version came out. Over the years, we've also developed a number of books completely in the open (a good example is the Python Cookbook online project we have done with ActiveState.) Things are different now, and there's a richer ecosystem for distributing downloadable-only versions. But I wouldn't consider this a copycat move. It's an obvious play, with lots of different variations.

You also said: "If O'Reilly were more proactive about getting the ball rolling on technologies that are just beginning to appear over the horizon..." Well, we actually have a pipeline model for that. We do online publishing on the O'Reilly Network sites, we do conferences, and books as a topic becomes more mature. It's pretty clear that we need to add downloadable PDFs to the pipeline, for topics that aren't ready for the retail channel yet, and as it turns out, we're doing that....

But I will say that we do sometimes feel the sting of being slow to develop titles we know are important. We spot things early, but we don't tend to rush out our titles, and as the industry has gotten more competitive, that's sometimes hurt us. A good example is Ajax. We co-hosted the original Ajax summit with Adaptive Path last May, and got several books started back then, so we were all over this new idea. But our first books are just appearing now, while Manning and APress have had books out for several months.

That is one aspect of the problem that David called diWorseification, otherwise known as too many balls in the air! Our authors had other projects. Brett McLaughlin had to finish editing Head First HTML and CSS before he could start writing Head Rush Ajax. (Head Rush is a derivative series we're starting, using the Head First methodology, but not promising to teach a topic from the ground up. Rather, it assumes pre-existing knowledge, in this case, of some amount of Javascript and XML. Defining the ground rules for that new type of book also took time.) Was that a mistake? Only time will tell. We like to think that sometimes it's worth not being first, if are going to bring something better to the table by waiting.

(Just checked: Ajax Hacks is now in bookstores; Head Rush Ajax should be there soon (both can be ordered on today); and Ajax Design Patterns is available in rough cuts format at .)

Hmmm...I hope that doesn't sound too much like an advertisement :-)

Challenge by Tom on March 25, 18:56

Thanks for the useful and thoughtful response, Tim. I didn't know about the JavaScript beta book; I stand corrected. Can't wait to see what rolls out of O'Reilly in the next few months!

Challenge by Aaron on March 25, 21:20

I heard 37signals pitched the Getting Real book to O'Reilly and O'R turned them down. I don't blame 37s for a second for doing it on their own after the traditional world said no (for whatever reason).

Challenge by Tom on March 25, 22:48

Because Getting Real is an entire book of fluff, presumably?

Kudos to 37s for finding a cute way to convert fame directly into profit: everyone wants to know The Secret Of Their Success, and is prepared to pay $19 if they think they'll get it. Unfortunately the book doesn't tell you what's actually made 37signals succeed: they're all really good at what they do, and they were in the right place at the right time (aka "lucky": see Microsoft).

They could have published a one-page document saying "Be really lucky, and hire people who are excellent at their jobs" and it wouldn't have been any less forehead-slappingly obvious; however, 5000 people would still have paid for it.

Challenge by FredS on March 26, 3:21

We get it - 37signals is the best and smartest company in the history of the universe.

Challenge by Gary Cornell on March 26, 5:07

A couple of comments from my point of view as the publisher of Apress and a pretty successful author as well. First off, I thought Tim's posts were generally excellent. As someone who also made the transition from successful author to (more or less) establishment publisher, I generally can echo what he said-though he said it far more eloquently than I could. In a nutshell (:-)), if all you do is publish bestsellers, the rules would be very different.

Still, I should begin by correcting one thing Tim said, Tim said that at Apress we started with a 50% royalty, that is not true and could be just a typo, our first royalty rate was 15% of our net across the board-or 50% higher than the industry standard but not 50%! Interestingly enough it is the rate we still pay once a book has sold more than 8,000 copies and the rate goes up even more as sales increase, ending at 20% of our net. This is how we at Apress share the profits with authors--with the most rapidly escalating royalty levels that we know of. However, it is true that we reduced our starting royalty rates to 10% of our net for the first 4,000 copies and 12.5% of our net for the next 4,000 copies because there was no other way to publish the specialized books that most other publishers have given up on. (Practical Common LISP is a good example, it too won a jolt award but isn't going to have sold anywhere near 10,000 copies any time soon. But the book sure is popular and well respected and I am honored to have published it.)

I do hasten to point out that we at Apress don't have anything to hide: our current contract is always available for all to see on our web site. So while it is true we have reduced the starting royalties for books that sell

Next, Tim is right about the level of sales, there is a reason that most publishers (except it seems Apress), are shrinking their programming lines and it isn’t because they are stupid, it’s because they weren’t making reasonable profits. Many of the more specialized, but we hope very interesting books, that we at Apress publish are ones that no other publisher seems interested in any more, they have very different sales patterns that our best selling Pro and Beginning titles. These specialized titles usually sell only a few thousand copies. Authors write these kinds of books to establish themselves, to help their consulting businesses, or their training businesses. If we paid 50% of our profits on this authors of these books they would probably end up with less than our current royalty rates! - but the authors are happy because the time spent on the book translates into higher consulting rates or more training gigs - and the money from those activities dwarf the profits that 95% of computer books earn. I used to charge $2500/day to do training for example and didn’t have trouble finding gigs—because of my books. Anyway, we at Apress are happy to publish these books because although the profits are small, they are profitable given our lower overheads then some of the other publishers and from my point of view they work for long term success. For example, they help confirm Apress’s new position as having the most comprehensive lines of books for programmers and IT professionals – a position we achieved in less than 7 years. We currently publish 10-12 books for this market a month more than any other publisher and more than Pragmatic has published in its lifetime I believe.

But it is a tough business, on average margins are small if you want to be comprehensive and fill all the niches that deserve to be filled and that dedicated people want filled.

Challenge by Diego on March 26, 6:01

Great article,
it's truly a sign of changing times.
In 2001 a friend of mine and myself were talking about online self publishing with pdf. We went as far as working out the logistics and I created a prototype of the site the would handle the whole thing for authors.
Everybody that we talked to looked at us like we had an arm growing out of our forehead.
Even recently I suggested it for a videogames magazine with super low budget and received similar treatment.

I am glad somebody else found the authors that believe in this method of publishing. Hopefully it will start something.


Challenge by bowerbird on March 26, 6:44

judging by the writing of the comments here,
everyone needs an editor except tim o'reilly...


Challenge by FredS on March 26, 7:25


Challenge by Chris on March 26, 12:07

I always wonder if it is possible for ALL of us to be creative publishers of things. Looking at the Myers-Briggs classification of personalities (Extroverted/Introverted, Sensing/Intuitive, Feeling/Thinking, Perceiving/Judging), it appears that many people prefer to be part of something than to actually create.

However, experimentation with agile methods of publishing is something that has not been done before, because of high transaction costs. Let's keep pushing new ideas and see what happens.

Cheers, Chris

Challenge by Harvard irving on March 26, 14:41

God, it's all such a bunch of shit. It's hard to believe that anyone is buying stuff from either self-publishing or the mainstream press. The tech world, and the literature world has become so faddish.

Does anyone have anything worth saying anymore?

Challenge by Kris Tuttle on March 26, 18:59

As far as editing or in a broader sense production is concerned, it is as important as the content. Some people may be their own best editors and producers but it is one in a million. This goes for everything including music. Thriller was 50% incredible production values.

The point I want to make is that there are different channels for this value add too. Great editors can be found out in the open doing freelance work not part of publishing houses. Bookstores are customizing their inventory to local markets and tastes and some even think like publishers as much as retailers.

What is so important here isn't that the old model is dead, there are many authors, books and consumers for which it will always work best. However there are also many that for the first time have much better options that fit better for what they are doing.

Challenge by Jake on March 27, 4:22

> Does anyone have anything worth saying anymore?

Harvard Irving certainly doesn't.

Challenge by Michael on March 27, 17:22

David makes some valid points about slef-publishing, but in the end I would have preferred (and doled out extra cash for) a traditional book.

For me the user experience of a physical book is better than screen reading, especially for longer reads. What was frustrating for me about Getting Real is that even though it was only available as a PDF, it was obvisouly laid out as a traditional book and there seemed to be no consideration for the PDF format. The table of contents was four pages long, with no hyperlinks. And it wasn't formatted for my printer, either. Using two columns a la a magazine layout would have preserved readability and at least 30% of the paper I used to print it out. My paper, my money. And then there were the all-black title pages in the original version (add my ink catridges. 37signals needs to apply its famous UI expertise to its next self-publishing venture. How about an e-book redesign?

Challenge by Robbie Allen on March 27, 18:51

Traditional publishers aren't going anywhere (for now), but things are certainly changing in the publishing world. Self-publishing and discount publishers are becoming more attractive options for authors. There is an opportunity for a new breed of internet publishers to correct many of the problems with print publishing.

In the internet publishing manifesto, I outline several advances I think are possible (and probable). It will be difficult for the traditional publishers to lead the way since their existing business models and practices get in the way.

Challenge by Todd Huss on March 27, 19:30

I have to 2nd Michael on "Getting Real" and that it prints out at over 160 pages with a very small amount of text on each page. I would have preferred a denser layout to reduce paper but really I wish I could have just bought a paper copy. That said, it's a great book as is Agile Development with Rails!

Challenge by Patrick Costello on March 28, 19:37

I think the "trick" to self-publishing, from my own experience at least, is to be willing to look at every available distribution method and find a balance that works for you and your project.

Publishing on the web is easy and the profit margin is great, but it's not a perfect solution because there is a group of potential customers who flat-out refuse to buy a downloadable book.

In some ways I can't blame them. As a writer and the chief cook and bottle washer of a small press I like the business side of electronic publishing, but at the same time there is something about reading a book on the computer screen that isn't quite satisfying. The overall experience of reading a book can't be duplicated onscreen - and while it's easy to say that experience isn't important in a technical manual you still can't discount the impact a positive customer experience has on sales. If it's data, shampoo or pornography a big part of the sale is always going to hinge on customer experience - and that experience involves everything from how the product looks on the shelf to how the customer interacts with it after it's bought.

A good editor is part of that end-user experience because the simple reality of writing is that it's almost impossible to objectively edit your own work.

In my own experience, the web wound up being a great sales tool. I write instructional manuals for banjo and guitar and our early experiments with PDF were disappointing. When I put the full contents of my books online under a Creative Commons license sales for the physical books suddenly took off. At first I was kind of confused because most of the people heading to Amazon, Borders or B&N to buy a copy of one of my books had already read the material written online for free. When I started asking people about that the reply was always something along the lines of wanting something to curl up with, share with a friend or carry around.

Even after they had the data they were still willing to pay for the experience of owning a physical copy of the book.

The other thing to keep in mind with self-publishing is that just putting the book out isn't enough. You have to draw your audience in - and in order to do that you have to be ready to do battle in the marketplace.

Let's face it, publishing something in a field dominated by a company along the lines of O'Reilly is no small undertaking. In order for people to choose your book in favor of something from a major publisher you have to literally build a new marketplace from scratch or run the risk of being overshadowed because a company like O'Reilly has more cash and more mojo to throw into an advertising campaign than a guy trying to hock one or two books.

Awards don't matter. Mentions in blogs don't matter. Reviews don't matter. The people who pay attention to those things won't buy your book. Don't let the popularity contest being waged on all levels of the web fool you into wasting time on the wrong marketing campaign. In order to get around the big players in the field you have to do things that nobody has thought of yet and reach the market nobody is paying attention to.

It's a complex business with bigger and more daunting problems than the choice of word processor. Before you buy into some Web 2.0 song and dance stop and look long and hard at what is really involved. There is no quick and easy answer and there is no one-stop solution because there are so many separate disciplines involved. To do it on your own you have to be a writer, designer, editor and marketer. You also have to deal with distribution, customer service, balance the books and be ready for the one thing you didn't think of to sneak around and bite you on the ass.

Challenge by Bill Bradford on March 28, 22:43

I've contributed to a number of Solaris certification study guides, and I *hated* having to write in Word (and use the publisher's templates) when I was writing about a UNIX operating system.. I ended up writing my content in vi, then cutting and pasting into their template before I sent the final work in.

Challenge by Alex Bunardzic on March 29, 2:17

To Patrick Costello: thank you for your advice. I think it is awesome.

I've also checked out your web site. Folk song of the Day is marvellous!

Challenge by chromatic on March 29, 4:18

Why do (some people) think Rails wasn't on O'Reilly's radar in late 2004? I don't think that it's a coincidence that people started to hear about it just as Curt Hibbs' first article came out on ONLamp.

Challenge by Ken Kousen on March 29, 9:09

I want to thank Tim, Gary, and of course DHH for a fascinating look inside the tech publishing business. Before I became a technical trainer I was a research scientist. In that field, you basically pay to have your papers published and the resulting journals still cost a fortune. I also spent a year trying to write fiction, which was an adventure (slightly over 100 rejection slips and a couple of very minor sales), but learned a lot about that business as well.

Some day I hope to write a highly successful book for one of you. Until then, I'm very glad for the opportunity to learn more about the process and what is needed to be successful. This has been a really fun thread to read!

Challenge by Computer book author on March 29, 9:17

Here we go again... "I wrote a book and sold it directly and it sold well and a traditional publisher couldn't have done it and I didn't have to use Word." Forgetting the specious argument about Word, this article says absolutely nothing useful. It's a niche book, for a niche group of readers, which will be forgotten in a year. The advantage of traditional publishers is, as Tim mentioned, the infrastructure they provide: professional editors, tech editors, proofreaders, indexers, sales reps, and all the rest that take a book from manuscript to bookseller.

Sure, your ruby on whatever book sold a few thousand copies. Now do it again. Then do it again. See what happens. Very few such initiatives work after the new-factor has worn off. (One notable exception is TidBITS' Take Control books.)

One poster points out the the ruby book needed an editor; so you're not providing readers with the best quality. Are you sure they'll all come back for more? What about when you have to update it? Let's see your sales figures then; how many readers will buy updates?

As for your numbers, well, Tim set you right on that. It's a shame that you write such an opinionated piece and can't bother getting the basic royalty numbers right.

Finally, when you write computer books, it's not to get rich, and it's not to get your name on the cover of a book: it's too make a living. There are many computer book writers (let's say 80%) who write one or two books, because they know the technology but don't really know how to write very well. Then the rest (20%?) know how to write, and they turn out books regularly. For those (that's my group, btw), writing books is part of an overall writing career that includes writing for magazines and writing for companies. Books help maintain cred; magazines pay well; and companies pay even better. There is a symbiosis among these different activities.

Good luck with your book. Make sure you post figures for the next one....

Challenge by Ken Kousen on March 29, 9:17

Incidentally, I used to avoid ebooks, too. Now, though, I almost always buy the pdf/hard-copy combos at Prag Prog, the rough cuts/hard-copy deals at O'Reilly (some of which are very, very rough -- see Rails Recipies), and the "early access"/hard-copy combos at Manning (which are only released a chapter at a time). I'm also a Safari subscriber at O'Reilly and through my university connections I follow Books 24x7.

I only wish I could read my ebooks on a plane when I'm traveling. In other words, I wish I had a battery that lasted the length of a normal flight. Instead I wound up spilling a soda on my copy of Agile Web Development with Rails, but Dave Thomas was nice enough to autograph it anyway. At least he knew I really was reading it. :)

Challenge by Computer book author on March 29, 9:17

Here we go again... "I wrote a book and sold it directly and it sold well and a traditional publisher couldn't have done it and I didn't have to use Word." Forgetting the specious argument about Word, this article says absolutely nothing useful. It's a niche book, for a niche group of readers, which will be forgotten in a year. The advantage of traditional publishers is, as Tim mentioned, the infrastructure they provide: professional editors, tech editors, proofreaders, indexers, sales reps, and all the rest that take a book from manuscript to bookseller.

Sure, your ruby on whatever book sold a few thousand copies. Now do it again. Then do it again. See what happens. Very few such initiatives work after the new-factor has worn off. (One notable exception is TidBITS' Take Control books.)

One poster points out the the ruby book needed an editor; so you're not providing readers with the best quality. Are you sure they'll all come back for more? What about when you have to update it? Let's see your sales figures then; how many readers will buy updates?

As for your numbers, well, Tim set you right on that. It's a shame that you write such an opinionated piece and can't bother getting the basic royalty numbers right.

Finally, when you write computer books, it's not to get rich, and it's not to get your name on the cover of a book: it's too make a living. There are many computer book writers (let's say 80%) who write one or two books, because they know the technology but don't really know how to write very well. Then the rest (20%?) know how to write, and they turn out books regularly. For those (that's my group, btw), writing books is part of an overall writing career that includes writing for magazines and writing for companies. Books help maintain cred; magazines pay well; and companies pay even better. There is a symbiosis among these different activities.

Good luck with your book. Make sure you post figures for the next one....

Challenge by Computer book author on March 29, 9:18

Sorry for the double posting - got an error message the first time...

Challenge by brianriggs on March 29, 12:09

If you are a middle aged American, who's tech-related job has recently become candidate for layoffs, and now are in school, learning new technology skills with hopes that it will add value to yourself(as an employee), you may not have a choice whether to buy traditional paper books versus the PDF versions of the same titles, simply because paper is heavy. Living in the current pro-globalization environment has nothing to do with having a bad back. Depending on what type of technology you're studying for, your bag can quickly total twenty pounds or more with just a few paper books. That, plus the two mile hike in the snow, from your Mom's basement to the bus stop, might result in a re-injury of your back, and a trip to the hospital, something that you could not afford at your age(with out checking the status of your medical coverage).

Challenge by Timothy Johnson on March 29, 12:40

I thought that the "Getting Real" book was at least a little bit of a letdown, just a conglomeration of all the blogs that I frequent daily. I was hoping for something a little more tangable honestly, along the lines of "Defensive design for the Web".

All in all, it left me with the impression that 37s was riding their wave of popularity to get a self-published book out there for some quick money. I remember them announcing their intentions to do it this route a while back, because of the poor royalties from their first book. I'm not saying that the book isn't useful, just that since I read 37s every day and David's blog here, there was very little new information. It is more like something I would try to sell inside my organization (which would have been a lot easier if it was a tangible book).

37s would not have been able to do this had it not been for their popularity from writing the first book in the traditional way. Admittedly, David has his own merits in creating Rails, and even if his fame wasn't tied to 37s, he might have been able to pull off something spendid.

It seems that from David's new vantagepoint, he has a unique perspective on what success is. That said, if we were all "hackers of the year", maybe that would be a hot year for us to self-publish and retain 90+% of the profits.

Challenge by Reader on March 29, 17:45

Ah, yet another youngster who thinks that a one-off success is "shaking up an industry"... Get some perspective - you obviously don't know much about publishing.

Challenge by Blake Schwendiman on March 30, 19:29

For what it's worth, I've published three technical books. The first was published by Osbourne/McGraw-Hill. The second two were published by me using

While I can't give exact numbers (because honestly I never could understand the royalty reports), the first book sold at least an order of magnitude better than my self-published books. Dollarwise, though, my earnings were similar for my first book and my second book (my third book didn't sell much at all -- poor topic or poor writing, I don't know).

So, my experience is that you can earn much more per unit using self publishing, but you lose a lot in marketing. Also, I believe that self publishing is much more likely to be successful for well-written non fiction and for technical books than for broad-market books or fiction. That's because marketing within one's specialty is likely easier to accomplish than marketing at large.

My forth book is fiction. I do not plan to self publish because I don't think I can do the marketing required to make it successful.

In summary, there's a place for both and there probably always will be. That's my $0.02.

Challenge by Ted Turner on March 30, 23:00

Now I'm just curious. I've use LaTeX a little bit. Do the pragmatic programmers use LaTeX or plain TeX?

Challenge by Dae San Hwang on March 31, 0:08

I realize that many of the commentors are discussing whether self-publishing is a viable option for the authors. I would like to say something from a reader's perspective on why self-published contents could sometimes be more valuable than a professionally published contents. (despite potentially lesser qualities some people are protesting)

I believe that large publishers optimize their businesses for volume sales. If it's obvious that a book will never sell more than 2,000 copies, it doesn't get published. (no matter how valuable a copy of that book will be to some small number of people) On the other hand, if a book gets sold more than 500,000 copies, it mean that a copy of that book is that much less valuable. Why? Because information's value is inversely proportional to the number of people who know it. Even though I don't buy a copy of that particular book, the topics will be covered in many places and many people will know them anyway that I will osmosis much of that information anyway.

While I've been studying programming for past several years, I find it increasingly more difficult to find books that cover what I want to learn. As my interests get more specialized, there are simply smaller market for the knowledge I need. I now turn heavily to blogs and news groups for the stuffs I'm interested in. I also find beta books, pdf only books very valuable because they tend to contain those (for the time being) niche topics I desperate want to learn.

I think self-publishing is more for the niche markets as David mentioned in one of the comments. Before the internet, self-publishing probably didn't make sense. It was too difficult to market your book to a smaller potential market. However, now it has become much easier to effectively penetrate that smaller market. (That is if your book indeed contains some valuable information for certain people.) I think self-publishing will possess some significant portion of the fat tails of future tech book publishing.

While some people argue that professional editing will make your writing more objective, objectivity is boring sometimes. When readers are likely similar to the author in some ways, (when the topic is highly specialized and niche) opinionated writing may be more fun and can transfer more than the information. It transfers the opinions and emotions of the author. I find them equally as educational as the information itself.

If your expertise is already established, (like David, for example) maybe you could even charge a premium for your book. Self-published book costing $200 for complete coverage of how to build, for example. I think I will buy it. Established book pricing scheme is pretty random, (from the buyer's perspective anyway) but I think niche books may even be able to demand the market price in some future.

By the way, don't you all agree that an average article on New York Times is worth a lot less (to you) than this blog article with all your comments attached? ;-)

Challenge by Daniel Read on April 01, 4:33

I've enjoyed very much reading the original post and this thread of comments. As a brand new entrant onto the software development book publishing scene, I find discussions of this sort endlessly fascinating. Thanks to David for getting the discussion started, and to Tim and Gary (and everyone else) for adding to it. I'll do my best to add something of value as well.

At the risk of appearing as if I'm using this as an opportunity to pitch my own venture, I would like to point out another direction that I hope to see more of in the future: small, independent publishers using digital printing and on-demand distribution for niche titles. This is direction we've decided to go with developer.* Books ( We are the very definition of an indie, with my girlfriend/partner and I working out of our homes, doing business totally over the internet and phone. We financed our first title ourselves, and it was just released a couple weeks ago: Software Conflict 2.0, which is a collection of essays about software development by Robert L. Glass, with a Guest Foreword by Pragmatic Programmer Andy Hunt.

Most books (including those by O'Reilly, Apress, and Pragmatic Bookshelf) are published using offset printing technology, which has been the standard for a long time. The strengths of this process are that it is extremely flexible (think of the huge variety of shapes, sizes, colors, bindings, and materials in the books you see in the store) and can produce a very high quality product. The primary downside is that the setup process for a print run is cumbersome, and it's only economical (given the publishing economics described so well by Tim above) to print books when you can print them thousands at a time.

In contrast, the "print on demand" model, as it's commonly known, uses a digital printing technology--basically a super-charged laser printer with book binding capabilities. The most obvious advantage is that there is no need to print thousands or even hundreds of copies of the book all at once. If I need a case of books to keep in my inventory for direct sales, then I order a box and they ship it to me. If a customer on Amazon orders one, they print one and ship it to that person.

And depending on the printer you use (we use Lightning Source, which is affiliate with the wholesaler Ingram) the books are of good quality. An experienced eye can tell the difference, but the average person can't. The cover stock is full color and a good weight, the paper is good quality acid-free paper, and the binding is strong.

There are disadvantages, naturally: your choice of cover stock is limited (no matte finish, for instance), you can't do color inside, and your ability to reproduce detailed, grayscaled graphics (like photographs) is limited. Also, the per-unit costs are higher, as you might imagine.

The primary disadvantage and challenge, though, is one of perception: because the vanity publishing industry jumped on the digital book printing bandwagon early and enthusiastically, "print on demand" in many people's minds is synonomous with "vanity/self published." This is a shame, and hopefully time will change this perception.

It's easy to identify a self-published or vanity-published book because the design will be amateurish and clunky. Even readers who can't articulate exactly why, they know instinctively that the book doesn't "look right." The cover looks a little cheesy somehow. The interior typography looks like it came out of Word with none of the margin or line height settings adjusted. There are typos, and there's no index.

The secret, we believe, to overcoming this perception barrier is high standards in title selection, editing, and design. Tim points out that one of the primary things that O'Reilly "brings to the table," so to speak, is strong and dilligent editing. Most people who have not been involved on the other side of book or magazine publishing have no idea the amount of editing effort that can go into even a mediocre book or article. And then you have design. For our first book we sought out an experienced designer who produced for us a design that I would put up against most big publishers software book designs. It took several months to get it right. Add in things like copy editing, tables and graphs, and a professional index and it's no trivial task to "crank out a book."

We strive to operate developer.* Books like any other scrappy independent book publisher, except that we use digital printing instead of offset. We don't just publish any old book because, even though the printing techology is different, it still takes a not insignificant amount of money and a whole lot of sweat to put out a book that's worth putting out--notice I didn't say "a book that will sell," because I have no idea how to do that yet, but a book that's worth putting out.

To see an example of a very successful company that looks and acts like a normal publishing company but that is using the print on demand model, check out Prime Books and its sister imprint Wildside Press. They publish an impressive range of high quality science fiction/fantasy/horror paperbacks and hardcovers. The books are all professionally designed and edited, and they are definitely *not* a vanity press--they're an indie. If you're hip to the sf/f/h fiction scene, you will be able to see that right off.

As experienced authors and publishers will tell you, sales come primarily from promotion and distribution. Promotion means taking out advertisments, paying for placement on, sending out review copies, sponsoring conferences, etc. All of that stuff costs money. Distribution in the traditional sense is about getting on bookstore shelves. If you have a good product, and you have a little luck, and you're willing to give a book distributor a ~70% discount on your book off of the cover price (retailers insist on a 55% discount, so the distributor needs a cut too), then you can take the risk of printing up several hundred or thousand copies of your book and sending them out "into the trade channel."

There is indeed a great deal of risk here (returns could be as high as 30-50%, and as Tim illustrated, you have to eat these), but the wider distribution (as Tim has also pointed out) gives you a real opportunity to scale up sales-wise.

Choosing Lightning Source was an important choice for us because of their affiliation with Ingram. Ingram will always show stock on our books (they call this "virtual stock," but it looks like real stock to retailers or anyone else placing an order through Ingram), which is key to being visible to the channel. If your book does not show stock at (in the US at least) Ingram or Baker & Taylor, your book for all intents and purposes does not exist to the channel--the friendly lady at your local indie bookstore might even be reluctant to order one for you if Ingram is not showing stock. So the point is that even though with this model you're not counting on traditional bookstore sales, you still have to consider distribution through the traditional channel--it's just more a pull *from* the channel (as people, hopefully, order your book) than a push *to* the channel.

One of the ways that we've decided to mitigate the distribution challenge is to focus on books that have a timeless quality to them, books that can stay in print for years, building up momentum and sales over time. This is a double-edged sword: we're not on the same technology-driven rollercoaster that heroes like O'Reilly, Apress, and Pragmatic Bookshelf are on, but we'll never make the kind of money they're making, and we'll never be putting out books on hot topics like Rails and AJAX. But hopefully we'll put out some books that we'll be proud of, books that will enrich people and be pleasing as physical objects, and hopefully the authors will make a little money too. Too soon to tell.

If anyone has any questions, feel free to get in touch at

Daniel Read
developer.* Books

Challenge by gmlk on April 02, 13:34

A niche is never found.
It's always defined by the one who carved it out.

Challenge by dmitry om. on April 04, 13:41

Personally, I prefer e-books by the following reasons:
* modern laptops are extremely convenient
and with you all the time
* I could transfer my pdf version to PDA and
read it in the forest
* it's not a problem to keep 50-100 books on
1Gb flash card and 10,000 on HDD.
I don't think I wish heavylift more than
2-3 paper books, with me.
What if I want to read couple chapters from
book about Ruby, then one from Rails, and
2-3 from book about SVN, for example?
* my computer assists me in search immensely
and does it much better and faster than I am
able to accomplish
* I have I'm electronic beta version now,
and I know that it's a work in progress,
and I, as a reader, could help to get
the final version better, providing feedback
in timely fashion.
* I don't care much about super-polished style.
After all, it's not a novel. I prefer to pick
up the brain of guru.
Technical books are different animals, and
sometimes code attached to one of the chapters
is more concise and clear interpretation of
the concept or ideas.

My paper book could be burnt, lost, stolen or look
clapped-out. My e-books are always new, and if I preferer, in any moment I could print out one or another chapter to make remarks.
If pdf was not protected, I could even make remarks inside e-book, too.

Paper books will die, earlier or later, I think.
Not during our lifetime, of course...

Challenge by João Marcus on April 04, 19:12

"modern laptops are extremely convenient and with you all the time"
That's because you don't live here, in Brazil, where laptops are too expensive for us. Programmers here don't get more than U$ 2000 a month. So, no, I don't prefer e-books.

Challenge by greg on April 05, 6:00

after reading tim o'reilly's defense of the tole of editors, I was reminded of a story well-known in business journalism about (now New York Mayor) Michael Bloomberg. seems that when Bloomberg was launching his business news operation on top of his already successful market data terminal, he turned to Matt Winkler, the guy who was charged with running the new operation, and said something like, 'if reporters do their jobs right, why do we need editors at all?' It's tempting to think that editors at newspapers, mags, book publishers, etc. are just correcting for weak writers, but the best writers in any business will invariably tell you about the one or two editors who help them do their best work.

Challenge by Bruce Eckel on April 06, 21:34

Here's a pretty good article that puts writing into perspective:

Challenge by Zebedee on April 10, 17:20

Question for Mr. O'Reilly: how many of your authors to whom you've paid hundreds of thousands or even millions of dollars over the years were *not* also the authors or major developers of the products they were writing about?

Questions for the publishers of Getting Real: how many libraries in the world carry this book? And if I purchase a copy, can I share it with a friend?

Challenge by DD on April 18, 5:10

so, i'm just wondering if this is a worthwhile endevour for would-be authors from a financial perspective. based on what i can glean from the above, expectations should be set for a $30 book on 10% to the author for about 50% of the list price. that would equal $1.50 per sold book. a seemily accurate volume of sales would be around 5000 books, based on the above, so this equals $7500 per book to an author. minus pessimistic personal taxes of 35% and an author gets $4875 per book. does that sound about right?

what is the burden time here to write one of these books? 80 hours, 160 hours, 320 hours? assume it is 160 hours, for the sake of argument. this equals about $30/hour for the labor. assume it's 320 hours and it's now $15/hour.

is this really worthwhile to even attempt in book writing? seems like an awful lot of blood, sweat and tears (and not to mention risk) for a paltry wage for an author.

but what i don't understand here is how this could even been anywhere near profitiable or appealing for a book publisher. assuming the above, a publisher would have to sell 100 books per year to get close to 1/2 million dollars in net revenue (not net profit). given the amount of staff, fixed costs, variable costs and taxes that it would take to run a publishing company, this is no where near a good business model.

am i missing something here in the calculations or value-proposition in book writing, from both an author and publisher perspective?

Comments are closed