Systemantics

daltonlp writes with the review below of John Gall's 1977 work Systemantics, writing "Most of the systems described by the author are societal or economic systems (governments, corporations, universities). Computer programs are mentioned, but they aren't the primary focus. But Systemantics doesn't distinguish between types of systems. In fact, its theories and arguments seem especially applicable to computer systems." (Read more below.) Systemantics author John Gall pages 111 publisher Quadrangle / The New York Times Book Company (1977) rating Insightful +5 reviewer Lloyd Dalton ISBN 0812906748 summary "A complex system that works is invariably found to have evolved from a simple system that works." Years ago, I saw this quote and committed it to memory. I've finally had the pleasure of reading the book it comes from. I was amazed that Systemantics was written in 1977. It's far more relevant today than it was then, because more people write more software today.

That means theories like
Systems in general work poorly or not at all.

Some might question whether this is really true for computer systems built with modern technology. After all, for a computer to function, millions of microscopic parts must act in perfect synchronicity at superhuman speed.

But in reality, computers fail much more frequently than we notice. A large chunk of their innards are dedicated to failing gracefully. There's ecc in just about every piece of hardware. Without it, computer hardware would fail too often to be usable. Software is no different--it can fail sooner or later, gracefully or catastrophically, but it's going to fail. Overall, computers work poorly, but they work.

Complex systems usually operate in failure mode.

In other words, something's always broken at any point in time. The measure of a complex system's quality is how drastically a particular failure impacts the rest of the system.

Loose systems last longer and work better.

Most Slashdot readers probably read the above and think either "Hallelujah!" or "Duh." But it's a small example of something I liked a lot about Systemantics. Buried under several layers of satire and pessimism is a genuine desire to help the reader avoid the mistakes of past systems designers and managers. There's more to this book than just pessimism.

What's Bad: Systemantics suffers a little from being a quarter-century old. Several references to Watergate and a few other cultural nods may be a bit lost on anyone under 40.

But the book's only real flaw is the author's occasional condescending tone. Every dozen pages or so, Gall takes the opportunity to criticize a real-world example. Some of these anecdotes serve as supporting evidence for an argument. Others are genuinely entertaining (the section on Job Goals and and Objectives is outstanding). But the author sometimes tries too hard to be satirical, and comes across as flat or patronizing, or departs on tangents unrelated to the book's central ideas.

Summary: Despite small imperfections, there's a wealth of real knowledge in this small volume. The author helpfully outlines the main points at the book's end (some of which I've bulleted above). The book's overall message couldn't be more clear if it summarized itself. Which it nicely does:

It is hardly necessary to state that the very first principle of Systems design is a negative one: Do it without a system if you can.

Systems are seductive. They promise to do a hard job faster, better, and more easily than you could do it by yourself. But if you set up a system, you are likely to find your time and effort now being consumed in the care and feeding of the system itself.

• New problems are created by its very presence.
• Once set up, it won't go away, it grows and encroaches.
• It begins to do strange and wonderful things.
• It breaks down in ways you never thought possible.
• It kicks back, gets in the way, and opposes its own proper function.
• Your own perspective becomes distorted by being in the system.
• You become anxious and push on it to make it work.

Eventually you come to believe that the misbegotten product it so grudgingly delivers is what you really wanted all the time. You are now a Systems-person.

You can find used copies of Systemantics from bn.com and other online sources, though good-condition copies fetch high prices. Slashdot welcomes readers' book reviews -- to submit a review for consideration, read the book review guidelines, then visit the submission page.

Comments

General Systemantics?

[Stile 65]

The name General Systemantics sounds an awful lot like General-Semantics, a theory of language and meaning that influenced Gregory Bateson, a cyberneticist and systems theorist. Is there any relation?

MOD PARENT DOWN. GOATSEX LINK + STOLEN TEXT

[thefirelane]

as others have pointed out

A true list of the problems of a small mind.

[Chemisor]

Only a small and untalented mind could come up with a pessimistic list like that. Only a man who has never made a working system in his life can hold beliefs like these. Beliefs that are self-perpetuating, contagious (though only to other small minds), and just plain wrong. Let me elaborate:

# Systems in general work poorly or not at all.

Only if the designer is a talentless slob or a commitee, which is all too often the case in the modern world. If your mind is clean and ordered, so will its products be. If your mind is a mess, full of nothing but superstitions, half-formed beliefs, prejudices, other people's opinions and bad advice, is it any surprize that it can not come up with anything that is good?

# New systems generate new problems.

Baloney. The problems are inevitably the same; there just aren't too many of them to begin with. Only applications vary, but to a small mind that is not capable of abstraction, each such application looks novel, untried, and unfamiliar. And what hope does he have of solving it in this particular incarnation? No more than in any of the previous ones.

# Systems operate by redistributing energy into different forms and into accumulations of different sizes.

A beautifully vague statement worthy of a beggar fortuneteller.

# Systems tend to grow, and as they grow, they encroach.

Systems created by small minds always grow because the mental process of abstraction is essential for their reduction. Look at any software project and you'll discover dozens of code passages that do exactly the same thing, but differ due to a haphazard and mindless design of the data structures.

# Complex systems exhibit unpredictable behavior.

If you create a system you should understand it. If you can not understand it, you should not create it, but delegate the task to someone who can.

# Complex systems tend to oppose their own proper function.

If you do not know what the system's proper function is (see previous point), how can you be surprized that your guesses prove wrong?

# People in systems do not do what the system says they are doing.

Because you, as the system architect, failed to understand its purpose, the people will naturally be forced outside it in order to accomplish it.

# A function performed by a larger system is not operationally identical to the function of the same name performed by a smaller system.

See the growth point above.

# The real world is whatever is reported to the system.

Garbage in, garbage out. Garbage without, garbage within.

# Systems attract systems people.

Systems attract stupid people who are too afraid to make a decision for which some system can not be blamed.

# The bigger the system, the narrower and more specialized the interface with individuals.

When you can not see the similarity between interfaces, they breed like rabbits. Sadly, each one eventually becomes attached to an individual who could be fired.

# A complex system cannot be "made" to work; it either works or it doesn't.

A perfect description of the mindset of a man who can not "make" anything work. Without abstraction and thought, information never becomes knowledge. Without knowledge, understanding is impossible. Without understanding, the man is a nothing but a pathetic blind, mute, and helpless cripple, created by his own listless hand.

# A simple system may or may not work.

A truly simple system always works. The hard part is making a system simple. Now that is the true talent.

# If a system is working, leave it alone.

And stick your head in the sand while you are at it.

# 15. A complex system that works is invariably found to have evolved from a simple system that works.

Bloated beyond recognition by clinging imbeciles whose only intelligence lay in avoiding screwing up the work of the person who created the simple system.

# Complex systems designed from scratch