author : Gerald M. Weinberg

This is Weinberg's introduction to General Systems Theory (reprinted 25 years after its original 1975 publication): a way of looking at, modelling, and understanding a complicated, complex world. This approach was deeply tied up with the cybernetics movement, which is currently reincarnating parts of itself as the complex-adaptive systems movement. It is worthwhile to learn as much from history as possible (something that often escapes many computer scientists, who seem to feel that the subject has no history), so I am going back and reading some of that movement's literature.

Weinberg goes through the principles, with a bunch of simple but illustrative examples, and some pertinent anecdotes. In the preface, he gives some quotations from extremely satisfied customers, countered by one from a computer programmer who says "I didn't learn anything ... It was a bunch of platitudes". I wouldn't go quite that far, but there was certainly nothing paradigm-shifting for me: the general computational modelling world has moved on in 25 years. However, I can see that for those who haven't already had a modelling-intensive education, there could be a lot of good stuff here.

It is most interesting in its final chapter, when it starts talking about taking a more process-oriented (as opposed to thing-oriented) view of the world, and references a book to come. I would like to read that book -- I don't think Weinberg got round to writing it.

This is a thoughtful and highly individualistic account of the art of influencing people at their request. (The reader is assumed to have made such a request, by virtue of the act of reading the book.) It covers a range of issues, including how to get information out of people, advice on running meetings, how to make the client not feel a failure for having to ask for advice, why solving the problem isn’t always the best idea, how to get people to take the advice they’ve paid for, how to get them to not need the same advice again, why not to get angry if someone steals your ideas, and how to set a fee rate. All this sound and pragmatic advice is illustrated with vivid little scenarios and anecdotes, from The Orange Juice Test, to the Principle of Least Regret, that should stick in the mind long after the book has gone back up on the shelf.

Every consultant, and everyone who hires a consultant, should read this.

In this slim book, Weinberg delivers his advice to aspiring technical writers, based on the metaphor of building a dry stone wall from “fieldstones”:

• Write only on what you are passionate about.
• Gather relevant pieces (the fieldstones) all the time, piled up ready for later use.
• When you want to write on a topic, select relevant fieldstones from your collection.
• Assemble them into the right order to make an essay, a report, or a book.
• Avoid writer’s block by controlling the number of fieldstones you are considering together.

This is all very sensible advice, and pithily delivered. Possibly the most interesting part of the book is where it goes all meta, as he demonstrates how to assemble the fieldstones into a sensible order by doing just that to assemble the description.

The fieldstone method is a pragmatic process for writing: you are doing it all the time, from gathering small snippets, to writing entire books. It won’t work for those times when you just have to write on a topic that you are not passionate about. Weinberg is privileged enough that this has not been a problem for him: he “cheated” at college to be allowed to write on his own topics, then in industry he was lucky enough that his boss took his first somewhat off topic report seriously, and now as a consultant he can write what pleases him. However, even in a non-voluntary scenario, if you have the stones, you can probably pull off an acceptable piece of work using this approach.