Rossdavidh's Reviews > Adventures of a Computational Explorer

Adventures of a Computational Explorer by Stephen Wolfram
Rate this book
Clear rating

's review

really liked it
bookshelves: white

I received a copy of this book in exchange for an honest review.

Halfway through this book, I was thinking it would be 3-stars, which for me is a typical rating. But by the end, it was 4-stars, which is well above average for me. Here is the story of why.

Stephen Wolfram is best known as the CEO of Wolfram, Inc, which makes software for scientific and mathematical analysis. Wolfram might object to that characterization as too narrow, which relates to the biggest problem with this book: it's written by a CEO.

CEO's, whether by the nature of their job or due to a selection effect or most likely due to a combination of the two, are untiring boosters. Their company is great, their company does amazing things, all the people that work at their company are awesome, their products are leading edge and they have been an innovator and leader in this space (whatever space that is). It gets a bit like the character "Topper" from Dilbert. The fact that Wolfram is, in spite of being a CEO, at all readable puts him way above average for his peer group. But, you can still see some evidence of this boosterism in his writing.

CEO's also have a lot of authority in their day job, and this might clash a bit with the need to allow someone else to edit their work. For example, Wolfram's previous book, "A New Kind Of Science", was over 1000 pages long, and partly as a result ranks up there with Stephen Hawking's "Brief History of Time" on the list of books which were bought way more often than finished. One example of this in the current work, is his repeated use of the phrase, "And yes, [x]". It starts to occur so often that it becomes distracting. An editor (of someone who wasn't their boss) might have done something about this.

However, there are plenty of positives to outweigh these factors. One is that Wolfram is obviously both a bright guy, and also one who is willing to think and say (and write) thoughts that have not been heard before. While you may or may not agree, for example, that his Principle of Computational Equivalence is a paradigm shift for science, it is certainly a useful and new concept to have in one's mental toolbox. I read this book a chapter at a time, usually in the coffeeshop in the morning before starting work. This allowed my subconscious to ruminate on each chapter for a day before sprinting on to the next. Lots of people describe books they really like as "a pageturner", but I prefer books which are so dense with ideas and food for thought, that you do better to digest each piece thoroughly before moving on. I like books that make me a tiny bit smarter, by giving me more concepts to use and patterns that I can recognize in whatever problems I am grappling with in my job and life. Wolfram shares a goodly number of useful ones here, and it was awfully decent of him to do so.

I also applaud his copious use of graphics. Charts, diagrams, pictures, screenshots, and more charts. When used well, visuals and text shed light on each other so that they are more than the sum of their parts, and Wolfram makes good use of visuals in this book. Some of them, it should be mentioned, were clearly made originally in color, and printed here in grayscale, but in the great majority of cases this was not a big issue.

One favorite chapter of mine was the one in which he takes a look at decades of data on his personal life, such as what he has typed, emailed, how his heartrate has varied, etc. He is mildly surprised to discover that 7% of all his keystrokes are the backspace key. He learns that his heartrate is significantly lower when outside, even after taking into account differences in activity level (e.g. comparing taking a walk outside to walking on a treadmill inside). This caused him to search for ways to do his (computer-bound) work outside, preferably while walking, a very computer-programmer kind of reaction to discovering this about himself. It is interesting to see both the numbers-and-data oriented side of him, and the human-with-a-life side, such as for example when he decides he is willing to travel to give talks in person, once his kids got old enough to enjoy going with him. Also, seeing a clear stripe of "no meeting time" in his personal schedule, across decades, that corresponds to dinner with family, is a nice touch.

I'm not sure if I agree that naming functions is like poetry, although I grant his basic point that it should be given a great deal of thought, and you have to make each syllable count for a lot. I do agree with his point that programming, especially in the design phase, is more or less applied philosophy, and this is not only because I have thought the same thing myself. There aren't many people who are programmers, who write about programming for an audience of non-programmers, and it is to Wolfram's credit that he does so. For something that is becoming more and more influential over the rest of society, how programming (and programmers) work is not very well understood by the rest of society.

I am unlikely to spend much time watching the livestreaming of company meetings which he has started doing, but it is an astonishing level of transparency that would be nice to see emulated. Imagine what the impact would be if Google or Facebook or Amazon were as transparent about what they are doing, and how and why.

Really, that's the final factor that pushed this up to 4 stars, and it comes back to the fact that he's a CEO, but this time as a positive. The CEO's of software/technology companies have become the objects of revulsion, fascination, speculation, and fear in the last decade, largely because of the fact that software has a bigger and bigger impact on people's lives, yet most people have not only no control over it but also little understanding of it, to even be able to predict where technology is going. Wolfram is doing more than any other living tech CEO I can think of, to try to communicate how software development works, and what he is actually trying to achieve with it. It would be good for our society, and perhaps for our tech behemoths as well, if they followed his example.
5 likes · flag

Sign into Goodreads to see if any of your friends have read Adventures of a Computational Explorer.
Sign In »

Reading Progress

October 20, 2019 – Started Reading
October 20, 2019 – Shelved
November 12, 2019 – Shelved as: white
November 12, 2019 – Finished Reading

No comments have been added yet.