Take a photo of a barcode or cover
A review by almartin
The Soul of a New Machine by Tracy Kidder
4.0
soul of a new machine is ~35 years into the history of computer design (after ENIAC, anyways); we're now ~40 years (!) past publication of the book. how does one even properly evaluate a 40-year-old contemporaneous history of hardware design & architecture? as a piece of history? for its utility in understanding instruction sets and machine code and computer design in the year 2022?
four stars as a text on craftsmanship and the joy of Making Shit Work; four stars as a ~biography of some engineers enduring outrageous toil to shift an incremental advance in the history of computation; three stars as an explanatory text about what, exactly, happens inside of a computer, and what the work of building one entails.
that might be heresy! it's hard to fault tracy kidder's for lacking a crystal ball - how was he supposed to know that [RISC](https://en.wikipedia.org/wiki/Berkeley_RISC) instruction sets were gonna take over the world? but it's also hard to write a 5 star review for a book that talks at length about microcode and instruction sets and how insanely challenging they are to debug when, like, a [conceptual revolution](https://en.wikipedia.org/wiki/Reduced_instruction_set_computer) ON THAT EXACT SAME TOPIC was happening contemporaneously?
like, at minimum, soul of a new machine needs a "20 40 years on" afterword that situates the book as the last gasp of a paradigm for computer design? idk, 4 stars, tracy kidder is a good writer and this won a nobel prize for a reason.
four stars as a text on craftsmanship and the joy of Making Shit Work; four stars as a ~biography of some engineers enduring outrageous toil to shift an incremental advance in the history of computation; three stars as an explanatory text about what, exactly, happens inside of a computer, and what the work of building one entails.
that might be heresy! it's hard to fault tracy kidder's for lacking a crystal ball - how was he supposed to know that [RISC](https://en.wikipedia.org/wiki/Berkeley_RISC) instruction sets were gonna take over the world? but it's also hard to write a 5 star review for a book that talks at length about microcode and instruction sets and how insanely challenging they are to debug when, like, a [conceptual revolution](https://en.wikipedia.org/wiki/Reduced_instruction_set_computer) ON THAT EXACT SAME TOPIC was happening contemporaneously?
like, at minimum, soul of a new machine needs a "