Well, that's a pleasant surprise. Somebody named Stephen has left a comment. And not one of those spam thingies, but some actual relevant questions about tech publishing. Answering them seemed like it might be fun..
1. I noticed there is a pdf version available of your books. Is this something you or the publisher insisted on? Have people shown an interest for this version or are they still overwhelmingly favouring the hard copy? Aren't you or the publisher afraid that the pdf will end up on shady torrents?
I honestly don't remember an explicit decision being made over whether or not an e-book edition would exist. As far back as I can remember, they wanted one and we wanted one, so no problem. I'd imagine that the publisher's wishes would override the author's in this respect if it came down to it.
I don't have the Manning contract for the wxPython book in front of me, but the Wiley contract for the Pro Rails book says "the Publisher shall publish the Work in such style and manner as the Publisher deems appropriate", and I think the Manning contract had similar language.
I'm in favor of offering e-books because they are easy to store, easy to update, and easy to search. From the publisher's standpoint, they have a relatively low cost to produce and people want to buy them, so it's a potentially large win.
As of the most recent statement I have on the wxPython book, e-book sales are just under 10% of total sales. I suspect that some publishers have much higher percentages -- Pragmatic, for example, really promotes their e-book sales. The Jython book is not available for sale in an e-book form, not even on O'Reilly's Safari Bookshelf. Which is weird, because they (at least used to) offer a competitors Jython book on there...
While I'm on the subject, I don't know right now if the Pro Rails book will be available via pdf. Wrox doesn't seem to do pdf much, and I don't know if they plan on starting.
Funny you should mention torrents, because I relatively recently got an email from Manning that they were trying to get the book removed from a torrent site. So I know it is, or at least has been, out there. I'd say Manning takes this kind of thing pretty seriously, but it obviously hasn't stopped them from providing pdf files. So far, I haven't seen any evidence that this kind of thing is affecting sales, and I'd be very surprised if such evidence presents itself. (Standard Disclaimer -- I don't speak for Manning, and I don't speak for Robin Dunn. If you broadcast a torrent of the book Manning will try and stop you.)
2. Technical books typically have a short shelf life and would probably benefit from frequent updates (e.g. incorporation of errata, version updates etc.). Wouldn't a more flexible way of publishing, with small, frequent runs (or even an outright on demand model a-la-Lulu) be better?
Yes, I could see where having small, frequent updates could be better for the book reader -- this is essentially what Pragmatic does with their e-books.
However, the infrastructure and overhead for doing this in printed books would be pretty significant, even with some kind of publish on demand model. At the very least, the cost to print a copy would go up, probably dramatically. Then you have the secondary problems of managing which errata went into which revision, continually updating the layout when needed, dealing with people who want printed books upgraded. The headaches get big quickly.
That said, something along that model is inevitable, at least for e-books. Eventually, other publishers will have processes as streamlined for this kind of thing as Pragmatic seems to.
This concludes todays episode of Ask A Guy Who Once Wrote A Book. Tune in next time. I hope there's a next time...