Steven Levy, in his 1984 book Hackers, spoke about computer nerds - a newly important class that had been charactered as weird outsiders only a decade before. Early in the book he invoked the MIT Tech Model Railroad Club as a cornerstone of the underlying culture.
The Model Railroad Club had two distinct group of members. One group liked to run the trains and the other was more taken with the underside of the table - all of the intricate wiring and control circuitry that could be created.
Engineering, at its core, is very much an art of compromise. Bits and pieces are brought together to create a piece of software or hardware that is as good as possible given a set of constraints that often compete with one another. Navigating these complexities is not straightforward and a very good solution may be crafted from subelements that are far from optimal by themselves.
It is fine and good if the engineer understands all, or at least most, of the constraints. The problem is that even excellent engineers are often too narrow to realize where and how their creations will be used. Within their speciality they may be very good at connecting the dots, but ultimately they may be ignoring the global picture. They can't see the forest for the trees.
All of us who are reflecting on Steve Jobs are running the risk of eulogizing him, but there is still something to be said. One of his skills, and of his team, is not only having a good sense of what technologies are available, but understanding why and how people will use their creations. They have a good sense of human centered design - something that is poorly known to most engineers.
An extremely telling glimpse of this came in the now famous 1997 Apple World Wide Developers Conference. Jobs was trying to assume Apple developers that there was a future in the platform. He was in the process of focusing Apple and that meant disbanding a large number of popular projects. Here he answers a critic. This is a beautifully crafted answer considering he is doing it in real time. It is also a viewpoint that fundamentally defines Apple.
Apple is not afraid to kill technologies as the company changes and grows. Some of those technologies may be much better than competing technologies in a local sense, but may not make sense in the final product. This totally frustrates a lot of engineers who often become fixated on some arguably great technologies.
Engineering is a great discipline. I’m not a member of the tribe, but I have great respect for the field. But engineering along is now insufficient in many product areas. As some classes of products mature they move beyond the point where improvements on the underside of the table lead to a more desirable product. Real attention must also be given to the other side of the table and the design should reflect the synthesis.
Some companies try to add human centered design by consulting with a few arguably great firms (IDEO comes to mind). Such consultations may be extremely helpful for many, but where the competition is ruthless and human centered design is important having to engage these companies is probably a sign of internal weakness and lack of focus. It is necessary to consider design very early on and there needs to be a solid bridge with the engineering side of the company throughout development.
I spent many years at a company that had an interesting human computer interaction department. It wasn't used directly and was mostly written off as a research expense. Their core expertise was impressive and could have easily coupled with human centered design and engineering, but there was little appetite for that sort of thing. The managers were focused ether on core engineering or spreadsheets and weren’t connecting the dots widely outside of their own experience.
I came to that department after having done a lot of experimental and applied physics as well as some conventional engineering. At the time it struck me that many of the problems of the next decade were solvable at the engineering level. The social level seemed much more difficult and became very interesting. I needed some schooling. It turned out to be much more interesting than I had imagined.
Apple is by no means perfect - perhaps more interesting than Apple's rise under Steve is how few have figured out there are two sides to the table and the piece not associated with traditional engineering is extremely important. It *really* isn't rocket science.
As an aside I think it is practical and useful to examine products and services and deduce something about the design process. This is something lost on a lot of tech reviewers who are often in love with certain technologies rather than the utility, functionality, design and elegance of a product - factors that may be much more important to the end user than processor speed, operating system, number of megapixels of a camera, or any one of a number of other metrics that are useful when refining individual dots.