In my Computer Vision class at MIT, our professor Antonio Torralba told us about a good book to read for beginning graduate students: Advice for a Young Investigator by Santiago Ramón y Cajal, a Spanish Nobel laureate who is one of the founders of the field of neuroscience.
While the book is clearly directed towards students at the end of the 19th century, it still contains a lot of advice that I think is still applicable today. Here is some of the key advice I found valuable:
- Do not be the mushroom around the tree of a great idea.
- Old ideas/theories of great people must be torn down for new ones to arise. Do not respect the famous people/researchers to much.
- Hunt for problems in established theories. Demonstrate them by hard data.
- Whenever a theory seems silly or not adequate write it down and think more about it.
- Get planned/scheduled time to think about the big picture.
- Seemingly small problems may just not be well understood.
- When you think about problems think about understood vs. poorly understood and not about how important the problem might be – this cannot be predicted that well anyway.