(from Keep the Joint Running blog )

Great Quotations: “My observation is that whenever one person is found adequate to the discharge of a duty by close application thereto, it is worse executed by two persons, and scarcely done at all if three or more are employed therein.” - George Washington.
-And thanks to David Mott for bringing this quote to our attention.


Generalists, according to the old joke, know nothing about everything, as opposed to specialists, who know everything about nothing.

Their managers know nothing about anything, which is why they hire consultants like me — people who know everything about everything.

Or at least know how to create that impression.

Smaller companies rely, for the most part, on generalists. Their receptionists figure out InDesign and use it to create brochures. IT doesn’t just code their websites — IT also organizes them and writes a lot of their copy, too.

Meanwhile, Fred in Accounting closes the books, sends out the invoices, pays the bills, and manages cash flow too.

And the owner variously serves as the top sales rep, lead product designer, head of customer service, and janitor, depending on what most needs doing and who called in sick that day.

Smaller companies have been populated by generalists since the days when the closest thing there was to a specialist was someone who was particularly good at chipping flint arrowheads, trading them for some mastodon steaks.

Bigger companies have relied on specialization at least since the dawn of the industrial revolution, when the invention of the factory resulted in jobs limited to a single task performed over and over again.

Employees’ sense of alienation and dehumanization, customers’ sense of being trapped in a bureaucracy, and management’s nostalgia for the days when business was fun … these accoutrements of excessive specialization are, I’m sure, just as old.

Why exactly do we do this to ourselves, when we all hate it so much?

One answer, I think, is being trapped in the mental habit of mass production.

Start with the difference between processes and practices . Processes are recipes: Follow the steps exactly and good results must follow. While practices also involve a series of steps, following them exactly leads to failure. Practitioners must apply their knowledge, experience and judgment in executing each step.

Great art is a practice. Turn it into a process and the result is a paint-by-numbers system. Being the carnelian specialist is unrewarding; the result is a parody of art.

Those who practice a craft take pride in the uniqueness of each item they produce. Their customers value uniqueness and are willing to pay more to get it.

Large manufacturing companies take pride in the opposite of uniqueness. They deliver enormous numbers of products as nearly identical as possible, built out of identical, interchangeable parts.

Assembled, of course, by fully fungible employees.

This industrial system works well at supporting the twin goals of throughput (high volume) and quality (adherence to specifications).

But business theorists extrapolated, applying the mass production model of increased specialization, organized sequencing, and interchangeability to business practices as well.

Which is how IT ended up practicing waterfall methodologies, convinced that if we just managed to become good enough at them, they’d suddenly start producing useful results.

No such luck.

In IT we’ve finally figured out that waterfall’s problems aren’t the result of a failure to perfect the process — they’re intrinsic to our having misapplied the mass-production model. Its specialization of roles adds overhead, impairs communication, delays delivery (it increases cycle time), and limits flexibility and richness of function (it reduces excellence).

We call the alternative “Agile.” It relies on generalists, who:

• Talk directly to business managers and end-users to figure out what they need, instead of working through an intermediary.
• Show business managers and end-users progress on a regular basis — often every day, instead of talking to them once during “requirements gathering” — making frequent design adjustments as a result.
• Write code, too.

When it comes to developing software, we’ve found that a collection of generalists, each of whom is competent in the whole craft, outperforms a team of specialists who segment responsibilities into a structured workflow.

The time has come for us to evangelize our discovery. Because elsewhere in the enterprise are managers who have been trying manfully (and sometimes womanfully) to force the square peg of factory-style organization into the round hole of the fast, creative, non-repetitive outcomes they’re responsible for.

Most managers, in fact, because very little of the work in a modern corporation looks like mass production unless your gaze is cockeyed.

When the work is more practice-like than process-like, mass production’s large number of hand-offs from one specialist to the next leads to results everyone hates — customers and managers just as much as employees.

I have an easy solution.

Let’s stop doing this.