As I read Kidder describe the toil undertaken in creating this new computer - working under the pressure on the brink of insanity to find those incessant bugs - I thought this the perfect companion for the CS154B Computer Architecture class at UC Davis.
Tracy Kidder, a journalist, not a computer engineer, took on the task of capturing the new computer building process when it was part science, part art, and to some extent part magic. We read about the vision for the new machine, the challenges of getting upper management to fund what they don't think can be done, the excruciating deadlines that must be met, the overwhelming frustrations during break-fix times, and the idiosyncrasies of the players.
He breaks down the complex technical nature of the task through a series of straightforward analogies and by doing so enables us to follow the human struggle of these mistreated engineers, who, driven by manipulation and pride work mad hours with little benefit to themselves to breathe life into a machine of their very own. 3. Everyone wants to feel an ownership of their project.
Consequently there are sections where the author takes great care to convey computer concepts and operations to a reader who has never seen a computer. As so many other reviewers have mentioned, the highlight of the book is the team and interpersonal dynamics--human drama that can be found in pressure cooker development environments today.
In desperation, a team in North Carolina was tasked with building a modern, 32-bit machine capable of addressing up to 4GB. In revenge, the passed-over team in Massachusetts decided they would build another 32-bit machine, on simpler lines, and show up the Carolinians.
While the book has sufficient technical information to keep the casual geek interested, the real story is about the engineers working 100 hour weeks and how this project became a big part of their lives.
The Soul of a New Machine describes the development process of Data General's Eclipse MV/8000, but Kidder has no particular insight into the industry or any specific aspect of hardware development. The reason this book was recommended to me was because the person who recommended it believes it gives a great impression of what this kind of deadline-bound creative development is (and, implicitly, should be) like, not just in hardware design, but in every sector that makes new things.
But the most interesting thing for me was that this was right around the time that marked the end of the ability for a single engineer to really know the entire CPU...they're getting into the territory where there just need to be some black boxes that you know about but know little of in order to get everything to work together.
Kidder may be best known, especially within the computing community, for his Pulitzer Prize-winning The Soul of a New Machine, an account of the development of Data General's Eclipse/MV minicomputer. The book was held to wide critical acclaim and became a New York Times bestseller.