Two disciplines of looking
Chess and drawing teach the same thing by two paths: recognising a configuration before calculating, and calculating when recognition is not enough.
Two disciplines have filled my evenings for a year, and they teach the same thing. In chess as in drawing, the beginner calculates everything, whereas the experienced player or draughtsman recognises first, then calculates where recognition is no longer enough. It is also the difference between reading a system and reconstructing it line by line.

A coach goes through games in front of students, saying out loud what to look at. The book passes on the reasoning that led to each move, with the candidate moves discarded and the reasons for each choice. It gives considerable space to quiet positions, where there is nothing special to find and a decision is still needed.

An autobiography useful for one specific reason: it shows the cost of preparation. Behind every admired intuition are years of memorised configurations, which the word talent is often used to gloss over.

Dated and dogmatic, it is still the best book on construction I know. It teaches laying down a figure in volumes and lines of movement before any outline, and preferring a clear reading to a flattering finish. The advice applies to an architecture diagram as well.