I've decided I ought to finally get around to finishing The ZooLib Cookbook, which I started quite a while ago and then got sidetracked away from.

My article on happiness is turning out to be much more difficult to write than I had anticipated, so I'm leaving it off for now to spend time thinking more deeply about the subject.

I've been wanting for quite some time to write a book, and figure that finishing the one I've already started would be the best way to go.

I approached a couple publishers about having them make a dead-tree edition of the zoolib cookbook, and even wrote up a pretty good prospectus, but got turned down by both because not enough people were using zoolib yet for the publisher to be confident they would sell enough copies to recoup their investment.

But I know the biggest complaint most people have about zoolib is that it's poorly documented. That's all my fault - I promised Andy Green I'd document ZooLib way back in 2000, and I haven't fulfilled my promise yet.

I think that if I actually completed the zoolib cookbook, it would encourage so many new developers that the chances of getting a dead-tree edition published would become pretty good.

I'm working at configuring my iBook both to support writing documents in docbook and to hold the zoolib source code and serve it to my other boxes. ZooLib being a cross-platform framework, what you want to do is put the source on a file server so that computers running other operating systems can get the same set of sources.

I have a Mac running Linux as a dedicated file server, but I want to be able to take my ibook to other places to work.

Update: DocBook

I could use some help from anyone who has successfully processed docbook XML documents on Mac OS X using the toolchain provided by Fink (a port of debian's apt package system to Mac OS X). I'm making progress, but right now I get hundreds of errors that are all like:

"X0174" is not a function name

I've gotten docbook to work on slackware and debian before, but it is always a chore, and I haven't found a nice way to do it yet that doesn't involve scripts with lots of hardcoded pathnames.