Occasionally, someone will pop up and say that Moose really needs a book. “X looks good, but it needs a book before I can learn it” is a common meme among programmers.

This is crazy, of course. Demanding a book before you learn something means you’ll never learn a lot of things. Book publishing is a risky proposition for many topics, and with the surfeit of good documentation on the net, it’s getting harder and harder to justify a book for any given topic. Even for books that aren’t failures, writing a book is not a good way for an author to make money.

I put a ridiculous amount of time into the Mason book, and my estimate is that I made $20 per hour (maybe less). Of course, having a book looks great on my resume, but the direct payoff is low. At this point in my career, it’d be hard to justify the effort required to produce another niche book, even assuming there was a publisher.

But the real point of this entry is to highlight just how much free documentation Moose has. A commenter on a previous post mentioned that he or she had created PDF output of the entire Moose manual. There are two versions at the link with different formatting, the shorter of which is about 58 pages. This is just the manual, not the cookbook. I imagine if the cookbook got the same treatment, we’d easily have 100+ pages of documentation. That doesn’t include the API docs, this is all stuff that might go in a book (concepts, recommendation, examples, recipes).

So please stop asking for a Moose book, and just read the one that already exists!

8 thoughts on “Moose Book? Well, Sort Of …

  1. Please note that the ridiculously low hourly wage for authors is, in large part, due to the ridiculously broken compensation model of most publishers. A book that sells for $40 may earn the author $2 for each copy sold and the publisher $14.

    Do publishers really contribute seven times as much value to the book as writers?

  2. Publishers do contribute some value. A lot of it is having a process in place for getting a book from idea to market. That’s less valuable these days than it used to be, with print on-demand.

    Anything a publisher can do, I can do on my own, but I’d have to front the costs (pay my own editor, pay for marketing, etc).

    But is that worth 7x? Probably not. I suppose if someone were to offer me a deal where I got $10 per book sold, I’d consider writing another book ;)

  3. Perhaps you should get those PDFs up on some on-demand book publisher (even CafePress does that these days).

  4. I think it’d be better to simply make generating nice PDFs part of the release process, so we could put them in the tarball and up on http://moose.perl.org

    That way people can print them out if they want, but the online version stays up to date.

  5. $10 per book is possible for a book with a cover price of $50. Any publisher who won’t pay you royalties of 20-25% of the cover price is cheating you.

    The standard rate is 5% of the cover price.

  6. Glad that you liked the pdf rendition of the manual. :-)

    Re: Manual + Cookbook. As a first pass, I only included the Manual::*
    sections, but nothing precludes me of adding the cookbook pods in the next
    edition. Or anyone else, for what matters: the recipe is now available on Github.

    Re: making manual creation a part of the distribution process. Ah, you
    anticipated my true end-goal. :-) Pod::Manual is still a very young and rough
    module, but once it matures a little bit, I have evil plans involving hacking
    support for it into Module::Build and co so that a fresh pdf version of the
    docs will never further than a ‘./Build manual’ away.

  7. I like the automation idea and I want a book(-let)!
    I’m even willing to work for it :-)

    I checked the LaTeX/pdf version. It’s a very good starting point.

    Dave, would you mind if I dropped the duplicate parts (AUTHOR, Copyright, …) and reduce this to just one occurrence ?

    I also would like to introduce more links into the pdf and I simply love a good index.

    The last two ideas would make it necessary to change the sources of the documentation, which I could prepare but Dave would have to redistribute the new versions afterwards.

    Is that acceptable?

    Wolfgang

Comments are closed.