Many Minds; What Work?

Two traditions of mind working are alive in the software industry. One is the tradition of the maverick: the solitary author alone with a blank page; the master engineer forcing a vision into reality through the efforts of lesser tradesmen; the enigmatic mathematician coding programs in back-to-front base-32 numbers. The other is the tradition of community mind work: the medieval university model of the community of scholars; the World War II model of the secret community of scientists; the 1970s model of the skunkworks.

The business world calls communities of mind workers teams, and devotes miles of print to them. Why? I think it's because the notion of the team meshes with three perceptions.

First, most people who are battling to survive in businesses feel bound by a view of competition that's too narrow to incorporate communal values, even when it recognises the need for them. Second, the communal structures business people know, not necessarily from personal experience, are the military, 60s communes, and 50s corporate families. None of those are effective when you need fast, coherent responsiveness to circumstances from many people. Third, from the wide interest in professional sports teams comes a sense that teams can be bought and sold, coached and controlled.

In business, the "team" sometimes has almost mystical connotations: a team exists independently of its individual members; team members willingly sacrifice their own interests in the cause of joint achievement; the value of the individual is affirmed by their place in the team. Somehow, these ideas then blur into notions that team members are "empowered", teams "share a vision", teams are walking, breathing embodiments of management jargon.

Muddle headed? I think so. Maybe that's why corporate teams often don't work.

Jon R. Katzenbach and Douglas K. Smith wrote a book called The Wisdom of Teams that distinguishes between working groups and teams, and further distinguishes between pseudo teams, real teams, and high performance teams. The difference between a team and "a mere group of people with a common assignment" is that a team has all these elements:

A team is a small number of people with complementary skills who are committed to a common purpose, performance goals, and approach for which they hold themselves mutually accountable.

That's quite a rigorous set of requirements, and most of the book is about what they mean, how they work together, and how to work effectively when they can't all be met. The authors are explicit that teams are not necessarily the best way for people to work together. Sometimes "mere working groups" are more productive, more effective, and more appropriate.

Katzenbach and Smith are also explicit that most productive, effective teams are not "high performance teams". Their definition of a high-performance team is that it "significantly outperforms all other like teams, and outperforms all reasonable expectations given its membership."

It's not obvious from the definition that enormous hours and huge emotional investment are essential to high-performing teams. In practice that seems to be what happens, which might be another reason high-performing teams are rare. But enormous hours, huge emotional investment, and fine products also happen outside teams.

Two books, published 15 years apart, are fly-on-the-wall documentaries of a year in the thick of computer industry projects. Both speak of long hours and deep commitment. Both end with the release of successful products. Neither reveals a team as Katzenbach and Smith define it.

In The Soul of a New Machine, Tracy Kidder describes a year alongside the designers and developers of a Data General computer. Their product, marketed as the Eclipse, was released in 1980. There's been a lot of 'human resources' theory under the bridge since those days, so it's fascinating to read this ancient story and recognise every element and most of the characters.

For I Sing the Body Electronic, Tom Moody followed the developers of a very different product in a very different company. The product was Explorapedia, an early multi-media encyclopedia, and the company was Microsoft. This story was published in 1995, but it's almost more fascinating in 1999, as the Web begins to define the mainstream. It has culture clash and personality clash, brilliance and bewilderment, despair and tragic irony: in this epic the heroes do not see their ultimate triumph. Although people talk often about the team and team playing, there's not much sign of common purpose or mutual accountability.

In figuring out how teams do work, Katzenbach and Smith have a fair bit to say about why they don't. One thing they don't say is that a rational response to a decade of downsizing, re-engineering, cost-cutting, and productivity extraction is to make no workplace commitment beyond the letter of a contract. That message does come through in The Self-Defeating Organization, by Robert E. Hardy and Randy Schwartz, another pair of management consultants who've distilled their experience into print.

The Wisdom of Teams reminds me of a peak 18 months in my working life. The Self-Defeating Organization lays out many of the other umpteen years. The only difference between these self-defeating organisations and the corporation of Dilbert is that Hardy and Schwartz acknowledge that even pointy-haired managers sometimes recognise the absurdity and internal incoherence of what they say and do. Management constraints can be just as real as engineering ones.

The Self-Defeating Organization is about organizations, mostly corporations, in general, but it does have an index entry for "team-building". Here's the reference:

For the sake of novelty or public relations, these shopworn alternatives may be introduced under trendy or euphemistic aliases. Thus an ill-considered retrenchment may be presented as downsizing; an ineffectual shuffling of the organizational deck as reengineering; a wholesale restriction of individual freedom and initiative as team building; or a program of systematic indoctrination and fear-mongering as retraining.

Depressing isn't it.

The Web might open the way to new working ways though. Open Source, its cultural cousin Shareware, and its dominant species Linux, suggest that working minds are figuring out their own communities, alongside and largely independent of corporate goals, focus groups, motivational factors, key performance indicators, corporate missions, and even teams.