Hacker News new | past | comments | ask | show | jobs | submit login
Richard Hamming: Learning to Learn (youtube.com)
174 points by sinwave on Feb 26, 2015 | hide | past | favorite | 16 comments



I watched through the series a couple of months ago. I will do my best to provide a couple of the highlights that I really liked. Unfortunately, I didn't keep a detailed journal of the course as a I watched. I really should have.

General Concept:

Richard Hamming is a Turing Award winning Computer Scientist and Mathematician. He spent a long and illustrious career at Bell Labs. As with many other promising scientists of the day, Hamming took part in the Nuclear Bomb effort and worked in Los Alamos. He refers to himself throughout the series as little more than a scientific janitor at Los Alamos. However, he realized during his time at Los Alamos that he had the rare opportunity to observe and interact with some of the most famous and brilliant scientists in the world. He decided that he wanted to join their ranks. However, after the war, he found that his effort was blocked by the fact that he didn't really know how to become great. There was no instruction course in become a great scientist. Thus, he set out to find the principles for himself. This series is his summary of the decades of investigation and personal experience in becoming a great scientist.

--------

Principles:

The following are a list of things that stood out in the series. Hamming had a number of principles that he constantly harped on. I will try to do my best to note some of them here and explain the context:

"Luck favors the prepared mind"

The longer you stay with a problem the higher the likelihood that you are going to find a creative moment somewhere in the mix. This idea reverberates throughout the entire series. In some sense, Hamming is trying to break apart the luck part from the preparation part of the equation throughout all of his talks. Interestingly enough, he declares that he doesn't believe that there is a formula for creativity (or at least a formula that any of us can use or know about). Instead, he harps on persistence as a part of this equation. You have to go after a problem for a long period of time. Once you have spent long enough with the problem that it has mades its way into your dreams, you know that you are approaching the point were rare epiphany moments are possible.

"With that which you learn from other you will follow, with that which you learned for yourself you will lead"

This is another quote that you hear time and time again throughout the series. By this he means that you actively have to work through practice problems and concepts yourself. He has found a vast difference in his understanding of topics that he passively soaked up and those subjects that he active dug into and made sure that he understood. In some sense, he used the Feynman technique. He pushed himself to being able to understand the concepts well enough that he could explain it fully and confidently to others.

"Find the people who are doing important things and help them do those important things."

The series is full of examples of thus. Hamming made it a habit to go after interesting problems with other people at Bell labs. He sought out opportunities to contribute to other people's work. It was clear the Claude Shannon was a genius, so Hamming made every effort to help Shannon out by providing computing assistance for Shannon's projects.

One of the fascinating habits that Hamming had was to grab lunch with different departments at Bell Labs. He would make friends but lose them quickly as well. He was constantly pursuing people with the question: "What are the big problems in your space?" He would then follow up with: "Well, why aren't you working on them?" This made people uncomfortable. However, Hamming made the decision that he was going to do 'grade-A' work, as he refers to it. Hence, he knew that he needed to associate with people who were doing grade-A work. If they weren't then those people were useless to him. He laments that fact that many brilliant people at Bell Labs went down as making no significant contribution even though they were far smarter than Hamming because, for what ever reason, were hesitant to tackle big problems with other people working on big problems. You don't get the feeling at all in the series that Hamming was apologetic for aggressively pursuing big problems. He wanted to do great work...period.

"If things are changing slow, listen to the expert. If things are changing rapidly, experts are only good for providing historical context."

This is one of the gems from the series. Hamming says that he used this as a rule of when to listen to established experts and when to ignore them in favor of the possibility for something else. I wouldn't say that Hamming advocated rejecting expert opinion in the situation that things are changing rapidly. Rather, he viewed their opinion as carrying just as much weight as everyone else's. One of the examples that Hamming brought up numerous times was the switch from analog to digital. He had a lot of co-workers who got stuck on analog computing. They were brilliant men who fought the the move to digital computing. The analog experts had countless reasons why digital computing would not succeed. So, Hamming tried to judge it by how quickly computing technology was changing. He realized that things were simply changing too fast for anyone to claim some sort of supreme expertise on the subject matter. Hence, he didn't discard the expert opinions, he took them as just another data point that weighed just as much as all the other data points. He didn't make any 'decisions' he simply refused to take a hard stand on the subject-matter until it was resolved through history. In other words, this principle is more about staying open to possibilities rather than rejecting them outright and putting all your chips in one basic based on an expert's opinion.

"In order to do the work that you want to do, you have to do it on your own time at the beginning. Only then will others give you time to do it."

This principle is quite possibly the one that arises more than any other through out the series. Hamming found it frustrating in his early years that they wouldn't let him research what he really wanted to research. It was a catch-22 situation. In order to show the importance of the work that he wanted to do, he would have already had to have done the research. Otherwise, his directors wouldn't give him the time to do so. One night he was reading a random magazine and realized that he was wasting his time that he could have otherwise been using to conduct the research that he wanted to conduct. Thus, from that moment, he decided to not waste any more time. No one was going to give him permission to start his desired research. He had to give himself permission and take the lead. He had to start by doing it in his free time. He cut magazines and TV out of his life. He spent less time with friends and on leisurely activities. Instead, he devoted his spare time to doing the research that he really wanted to be doing. In the end, he was able to do enough in his spare time that he took it to his directors at Bell Labs and they then gave him the time to continue pursuing the research on the company dime. However, he didn't stop there. He took that opportunity to fill his newly spared time with other projects. He kept it up for decades.

"Stay abreast of new things"

There were so many new things arising everyday in computing that it felt overwhelming for Hamming. However, he knew that he needed to stay aware of what was happening. He didn't take that to mean that he needed to become an expert in everything but he needed to be aware of it. Hence, a lot of his spare moments in the day were filled reading about new developments and talking with people who were working on interesting projects.

"Friday afternoons were for big ideas"

Hamming had the habit of setting aside friday afternoons for thinking about big ideas. He was religious about this time. He would avoid everything else that he could so that he would have friday afternoons free to think about where the future might lead.

I wish I had the time to pick out more principles that I gleaned from watching the series. There are countless nuggets throughout the series. He spends a lot of time talking about the details of the different projects that he worked on. However, those are nothing but stories to frame the principles that he is trying to communicate. The above principles that I have listed at only a few examples. His stories certainly help give more context and make it easier to remember the principles. It would be easy to write a full commentary on his lectures.

EDIT: I've added a couple more principles that I remembered. However, nothing replaces actually hearing Hamming tell it in his lectures. I watched them at 2x speed on YouTube. It is easy to follow even at that speed. Hamming speaks relatively slowly at normal speed. So, 2x speed just feels like a quick conversational pace.


Hamming did not win a Nobel Prize. Perhaps you meant the Turing Award?


Yes, I did. Sorry about the mistake.


Thanks for your introduction! Are the videos essential in that there are many diagrams and visualisations shown or can you get most or all of the meaning by listening to the audio portion and reading through the chapters of the book?


Listening to the audio will certainly do the trick. He runs through a lot of the big picture explanations of the problems that he was working on with the whiteboard but those aren't primary. By his own telling, all of the problem details are really just to frame the principles that he is communicating.

That being said, I watched with the video. I didn't try to only listen to the audio. So, it might be harder to follow without the visual aspect. And, he randomly pops in the principles. I can't really say that the highlights were only at the beginning or ending of lectures.


In case anyone is interested in a tl;dr, PLOS Computational Biology just published a "Ten Rules" article on Hamming's talk: "Ten Simple Rules for Lifelong Learning, According to Hamming"

http://journals.plos.org/ploscompbiol/article?id=10.1371%2Fj...


I think that genius takes extraordinary intellect but also a penchant for timing. Solve interesting problems, but also at the right time.

A way to overcome this is the Erdos approach: have a backlog of problems in your mind and allowed them to be solved naturally as you learn and new discoveries are made. Maybe that is why he was such a prolific mathematician.


One of my motivations in building videos for: https://www.youtube.com/user/devfactor

Was just this. In school you can listen, and be taught. But learning how to learn is an entirely different skill. Foundational knowledge is really important, but eventually you need to know how to pick up things on your own.

Understanding the learning process, and the process of seeking out relevant information is almost more important than being able to memorize text and lectures. If you can only be taught, that you rely on schooling. If you can learn, than your limits are only based on time :)


Can someone who already watched this comment? Anything particularly great about it?


I haven't watched it either, and would like the same answer.

The book on the topic is at http://worrydream.com/refs/Hamming-TheArtOfDoingScienceAndEn... .

I read Hamming's "You and Your Research" some years back, and recall that it was quite appropriate. See http://www.cs.utexas.edu/~dahlin/bookshelf/hamming.html for the text. It's the penultimate lecture in the video series.


Thank you! And I can't help but notice a couple more interesting titles in that directory: http://worrydream.com/refs/


Ooo! That's quite the collection of classic works.


It's a pretty good course on how to be not merely technically skilled, but effective. For example, how technological progress usually works and the importance of back of the envelope calculations. He teaches with a lot of personal narratives which are easier to relate to and remember.

The book is easier to read than watching the videos, though you could rip the audio and it would be an okay audio book.


I have read his "you and your research", and it is a really good essay, pg has it on his site. http://www.paulgraham.com/hamming.html


I would really appreciate a short summary to get a few more of the basic ideas taught in this class or book.


Fantastic! Thank you for this.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: