Johnson unveils some key dynamics of innovations and offers critical assessment about the history of technological advance. Hence this book benefits those who want to study this field. But Johnson himself is, nonetheless, an outsider from the research profession. His several arguments are misleading and further evidences will cast doubt on his conclusions.
The most misleading thing is his illusion that research is a directed activity that mimics evolutionary process. Such argument makes no sense to those Phds who spend >4 years in studying on the same subject and progressively narrow their visions. Johnson believes research is mostly about combining the "adjacent possible" and using "platform" and "networks" to arrive at transformative conclusions. This idea just ignores the fundamental forces that have science so powerful: the mathematical and logic rigor.
Johnson's misconception is due to his ignorance about what is different between TODAY's science and the science before 1930s. Perhaps, Johnson's arguments apply to those old-time elite scientists who spend their time in coffeehouse and brainstorm with other peers. This nostalgia no longer exists. Today research largely becomes a business. Since it is a business, you need to study the most pertinent thing about business - incentives.
The lack of an analysis about incentives makes Johnson's book incomplete. Without understanding how this decisive incentive thing drives different scientists to perform differently, we can know little about why some countries are more innovative (i.e. furnishing more talented researchers) than others. We cannot comprehend why innovation ACTIVELY takes place somewhere sometime.