How to read a technical book/textbook

I have only come to accept this after accepting limits on my time
There are no cookies for "finishing" a textbook
This is very different from reading fiction; it's much more "active" – even "interactive"
What I want from the book is not necessarily what the author wanted to teach
- Potentially antagonistic
- Compared to fiction where I actively suspend disbelief and follow the author

Technique
- Clarify intent
  - what do I want to know?
    - Learn to apply a specific topic
    - Do a generalized survey
  - perhaps I'm not sure what I want to learn
    - start skimming several books/papers
  - determine budget in time and energy
- Choose a book to read
  - Look for reviews, lindy effect
    - Don't spend too much time here
  - Rely on people I trust for recs
- Actively skim the book
  - Read the ToC, index, blurbs, diagrams
  - Any pages that stand out
  - Don't stop on hard things
  - Don't accept the book as true; don't fall for suspension of disbelief
- Mark up the book
  - if I know what I wanted, highlight those pieces
  - if I don't know what I wanted, skim first to figure out what I can get
    - then mark them up
  - brief summary from skimming, and decide what to do next
- Engaging with the book
  - Confirm I want/need more informaton/knowledge
  - Decide how to judge that I've acquired what I needed
    - project? notes? article?
    - exercises?
  - Rely on skimming to figure out how I'll tackle the book
    - Where I'll pay attention, what I can skip
    - What other books/references I should pull in
      
      Are there other references that will match my purpose better?
  - Read the book, paraphrase it constantly
  - Implement projects based on the book
- Consolidating the knowledge
  - Write up summary in my own words
  - Try to explain it to someone directly & simply
  - Maintain a list of Jargon / Glossary for easy look up
References
- How to read a book, Adler/Doren
- How to read a book, Paul Edwards: http://pne.people.si.umich.edu/PDF/howtoread.pdf