View profile

Seeing like the TSA; Technometria - Issue #23

Seeing like the TSA; Technometria - Issue #23
By Phil Windley • Issue #23 • View online
Corporate and hierarchical, bureaucracies strive for legibility. But people aren’t corporate or hierarchical. Instead they’re messy and innovative.

I just flew for the first time in 18 months. In that time, Salt Lake International Airport got a new terminal, including an update to the TSA screening area. The new screening area has been touted as a model of efficiency, featuring bin stations for people to load their bags, electronics, belts, and shoes into bins that they then push onto a conveyor. The bins are handled automatically and everything is sunshine and joy. Except it isn’t.
The new system is perfect so long as the people using it are too. The first problem is that unless your at the last bin station, the conveyor in front of you is constantly full and it’s hard to get your bin onto the conveyor. And if you’ve got more than one bin to load, they are separated from each other because the loading station isn’t big enough for two. People just don’t conform to the TSA’s ideal!
But the real problem is that people forget things in their pockets or don’t take off their belt. In the olden days, the TSA had little bowls. You’d throw your stuff in one, put it on the belt, and be on your way. Now, there’s no easy way to accommodate forgotten things except to go back to a bin loading station and put them in a big bin, clogging the conveyor even more. Three people in line ahead of me at the scanner forgot something, causing all kinds of delays. The TSA people were even telling them to just hold them in the scanner and taking them from them to hold while the scan was completed. Because there’s no good way to deal with forgotten items, everyone is forced to improvise, but the system is rigid and doesn’t easily accommodate improvisation.
The situation reminded me of the story James C. Scott tells in the opening of Seeing Like a State where forestry officials planted neat, efficient rows of trees instead of letting the forest take its natural path. The end result was less yield from the forest, but happier foresters who could now see every tree. Scott’s point is that bureaucracy aims for legibility in order to serve its own purposes—and usually fails in that effort. The primary reason states want legibility of citizens is taxes (and historically conscription). The TSA has ordered the screening process and made it legible to the screeners, but made no provision for outliers. If no one forgets anything and the system is lightly loaded, it should work great. Of course, that’s not the real world.
IT people are bureaucrats in their own way. We build and operate the systems that people use to do their jobs and live their lives. We strive for legibility in order to make the software simpler for us, even if it doesn’t serve the users quite as well. Universities are decentralized places with lots of innovative people pursuing their own goals. They are more feudal than corporate. I’ve often heard university IT people complain about this reality because it makes their life harder. If you’re a professor, you’d like to use whichever LMS suits your particular needs. But that’s not very legible. If you’re a university IT person, you’d like to force all faculty to use the standard LMS that the university chose. Neat and orderly, but it squeezes the innovation out of the university one drop at a time.
Life is messy. People are forgetful, disorganized, and, relatedly, innovative. Bureaucracy desperately wants legibility so that the rules are followed, the processes perform, and the bureaucrat’s life is made easy. Building systems that support decentralized workflows and individual decisions, without getting in the way, is hard. And letting people be people can be frustrating when it’s causing you headaches. We’ll never build systems that support an authentic, operationalized digital existence until we stop trying to fit people’s decentralized lives into our neat, ordered, legible software.
Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed James C. Scott
End Notes
That’s all for this week. Thanks for reading.
Please follow me on Twitter.
If you enjoyed this, please consider sharing it with a friend or twenty. Just forward this email, or point them at my news page.
I’d love to hear what you enjoyed and what you’d like to see more (or less) of. And if you see something you think I’d enjoy, let me know. Just reply to this email.
P.S. You may be receiving this email because you signed up for my Substack. If you’re not interested, simply unsubscribe.
© 2021 Phillip J. Windley. Some rights reserved. Technometria is a trademark of PJW LC.
Did you enjoy this issue?
Phil Windley

I build things; I write code; I void warranties

In order to unsubscribe, click here.
If you were forwarded this newsletter and you like it, you can subscribe here.
Powered by Revue