Backwards induction is one of the most powerful and underutilized scientific techniques. It is incredibly common that you can't prove some proposition A is true, but you can always assume that it is true and determine what that implies. By following the chain of implications you can sometimes make predictions that are provable, which then supports your hypothesis.
Would like to know how quickly Shannon cycled through these steps. The "fill your mental matrix" may suggest some for of priming or unconscious process taking place, perhaps over night, or over a walk? My experience is that ideas take time to bloom.
Nice list of tricks. I always try to remember one simple truth of problem solving and creativity: our minds can create nothing. All we can do is find connections between things. We consider it genius when people make a great leap of a connection, but that is usually achieved by the hard work of finding the right inputs. Shannon understood that as well as anyone.
Putting Claude Shannon in an ordinary grad student's shoes, what fills me with sheer unmitigated terror is: if that's what was my masters thesis, what could I possibly do after that, that is significant in comparison, that would qualify for a PhD dissertation. Stuff for nightmares and cold sweat. Surely not for the real Claude Shannon
I was surprised at the recent surge in articles about Claude Shannon. Taking a closer look, I can see a brilliant case of how good marketing works. Most of these articles have been written by same authors who wrote the biography of Claude Shannon. Although, he is a respected figure but his reverence—or, better say, popularity—is nowhere near a famous businessman like Steve Jobs.
These articles work as a trigger to get interested in Claude Shannon and then, the book itself. And I am sure it's working. I am almost tempted to buy the book. Like many people, I certainly detest many marketing tactics but this is an example of one that has been done right and is even helpful.
Also, I got an email because at one point I had tweeted about Shannon:
Dear Sam,
I hope this note finds you well. I noticed that you had tweeted in the past about James Gleick's book "The Information." As it happens, that book inspired a book I've just finished: the first-ever full length biography of the late Bell Labs engineer Claude Shannon.
The book was recently published by Simon & Schuster. I figured I would reach out, given your interest in Bell Labs. We were fortunate to have worked with Alice Mayhew, the editor behind A Beautiful Mind, Walter Isaacson's Steve Jobs, and other books in that genre.
I think you'll enjoy the book and hope you get a chance to check it out!
Jimmy
--
--
My new book, A Mind At Play (Simon & Schuster), is available now. It's the story of Claude Shannon, one of the founders of the information revolution—and one of the reasons we can exchange these emails.
Causation could perfectly be the other way around: Some people write a book about Shannon, and thus get asked by publishers to write articles about him for them. Plus you wouldn't expect the authors of articles about someone to be people that know little about that person.
It's hard to be involved with programming for a long period of time and not run into Shannon, he is one of the foundational people in who worked on the theories underpinning modern computers and telecommunications.
Another is Dijkstra,
In fact it's not overstating the case to say that Dijkstra's algorithmn (and it's descendants) and the Shannon limit are two fundamentals to the way the internet works.
I learned about Shannon about a year and a half ago. Very impressive figure, but it was clear that he wasn't in the popular consciousness at all. I have noticed a significant increase in the number of things being put out about him and had been wondering why. I'm not surprised a concerted effort is behind it. We're all the better for learning about him though so I welcome it.
Step 2: Fill your 'mental matrix' with solutions to similar problems.
Step 3: Approach the problem from many different angles.
Step 4: Break a big problem down into small pieces.
Step 5: Solve the problem 'backwards.'
Step 6: If you've solved the problem, extend that solution out as far as it will go.