Errors in Conceptual Design
Look to past failures to unravel how technical design is intertwined with economic and aesthetic objectives.
Learning from the Poor Judgment of Paconius
In “Design Paradigms: Case Histories of Error and Judgment in Engineering,” Henry Petroski details several civil engineering failures throughout history. The text also includes candid analysis of the reasoning that led to each failure. Petroski begins his discussion of errors in conceptual design with the story of Paconius and the Pedestal of Apollo.
Let’s explore how this civil engineering lesson can be applied to the design and development of software. If you’re not familiar with the story, listen to the radio program linked above or read this transcript.
Paconius sought an innovative solution to moving a massive pedestal from a quarry to the building site at the temple of Apollo at Selinous. The factors that contributed to his design appear to be primarily economic and aesthetic in hindsight. His solution, rather like a spool unwinding from the top, was apparently easier to construct, and allowed for heavier loads to be pulled with fewer oxen. Moreover, in contrast to the standard wooden frame solution, Paconius’ innovation required no holes to be drilled in either end of the pedestal (to be fitted with iron and lead inserts). The spool approach also kept the pedestal more protected from the road and elements.
These all seem like big wins. Therein lies the problem, it’s a trap, and a seductive one at that. There are two fundamental flaws with Paconius’ spool:
- As the oxen pull the spool forward, they distance themselves equally from the spool.
- After the rope runs out, you have to set the whole damn thing back up.
As the oxen distanced themselves from the spool, it became less stable and was prone to running off the road, undercutting the economic efficiency of using fewer oxen. The problem of scaling the solution over long distances made things even worse for Paconius. He ended up in financial ruin.
The Folly of Competing Objectives
Every consumer facing software project has economic and aesthetic objectives that are weighted to varying degrees. How do we balance these aims with technical design?
There are clearly many examples of software projects run off course, similar to Paconius’ failed endeavour. However, I’m going to focus on how technical, economic and aesthetic objectives can be acheived harmoniously per an article I read on “Designing for Performance,” by P.J. Onori.
“Designing for performance requires a commitment to maximize the necessities and strip out superfluous luxuries. A more utilitarian approach towards design is needed where each element has a tangible and net-positive impact on the end result. This does not negate the importance of aesthetics, rather it means that it should not given a blank check.”
No blank checks doesn’t mean letting the engineering team make all the decisions, it means regardless of your role, embrace an engineer’s mindset. One reason why VCs are insistent on startups having technical founders is their ability to assess the implications of conceptual design. Think carefully about how early design decisions impact technical architecture and performance. At the end of the day, we’re all trying to solve problems, and this leads into my favorite part of P.J.’s article:
“…aesthetics should always support the problems we are asked to solve…. Looking at design through this lens will help create a goal-driven approach and focus our thinking.”
In Summary
Goals alone aren’t enough to keep a project on track. To spare future generations the humor of reading about your own spectacular failures:
- Recognize competing objectives and resolve them as early as possible.
- Explore novel problems and potential solutions with users, designers and engineers.
- Question and attack your own ideas, but don’t stop having them.
Originally published at www.rockhopperdigital.com.
