Agile Careers

 

Jim ("Cope") Coplien is an old C++ shark who now integrates the technological and human sides of the software business as an author, coach, trainer, and executive consultant. He is one of the founders of the software pattern discipline, and his organizational patterns work is one of the foundations of both Scrum and XP. He currently works for Gertrud & Cope, is based in Denmark, and is a partner in the Scrum Foundation. He has authored or co-authored many books, including the recently released Wiley title, Lean Architecture for Agile Software Development. When he grows up, he wants to be an anthropologist. 

Log in to register a comment.

Subscribe here.

Blogs Blogs
« Back

Autonomation

A ghost from my childhood still haunts me: the memory of luddites fearfully pitted against factory assembly lines. What’s odd is that in spite of the familiarity of the rant from my childhood, I can’t recall the last time I heard it on the street or read it in the news. Automation is an accepted part of modern society. Perhaps it stands on the shoulders of the 1960s Utopian hopes that technology would relieve us of mundane tasks and give us lives of leisure, or perhaps society is satiated that demand has just caught up with supply.

We usually associate automation with industrialism, but the arts aren’t immune from technology. This is an important side discussion because engineering lies in an uncomfortable space between science and art. Modern art history has been a struggle between the forces of mass production and human individuality: the beaux arts, the Arts and Crafts movement and Art Nouveau celebrate the human in art; the London Great Exhibition of 1851 and Art Deco highlighted the contributions of machines. Most modernistic schools have this same industrial streak.

In late-19th century Japan Sakichi Toyoda noticed that his looms wasted cloth when continuing to run after their yarn ran out or when weft threads broke. His inventions put “human intelligence into the machine.” The looms could now automatically detect these conditions and shut down so that a person could repair the problem, restart the loom, and avoid the waste. 

Taiichi Ohno would later incorporate this concept as a pillar of the Toyota Production System, and called it autonomation. It would appear in Toyota plants as, among other practices, “stop the line.”

Ohno distinguishes autonomation from automation as having a human touch or a limited form of human intelligence. Autonomation doesn’t fix problems: it merely stops the machines so a human can fix them. Full automation requires that a machine find and fix its own problems, but that’s rarely cost-effective. Let humans intervene for the quirky job of problem-solving rather than doing the dehumanizing job of standing guard over the assembly line. While full automation is not cost-effective, autonomation is.

Computers are heralded for automating our tasks. Why does it not give great joy to employees that use electronic time-card systems instead of paper? Because the paper never worked in the first place. Autonomation not only goes above blind automation to accommodate errors, but focuses on techniques that have been proven to work without machines. Too much software today promises only to do things differently than we have done in the past. This is the same misplaced assumption that fueled modernism in design: that to be new or different was automatically to be better, if it employed the latest technology. Give a graduate class the assignment to create a tool to help a team track progress on its tasks and they’ll computerize the task: No CS student will consider solving the problem with stickies on the wall, let alone consider evaluating whether that might be a better solution than the automated one.

Too much software today pushes users into a position of meeting the machine on its level and its terms, forcing people to do things that machines do well. The programs aren’t intentionally masochistic: they’re striving towards full automation, but they fail with a high cost to the end user experience. Consider that the next time you submit an ACM paper online or use another online web service.

Prove in the human processes first, leaving out automation, and then automate that new process if it helps. Avoid technology for its own sake.

Comments
Trackback URL: