New Blog

To Don Knuth: teacher, algorithm analyzer, and inventor of TEX

So reads the dedication to the textbook we use in my Algorithms class.

Dr. Purdom frequently drops Knuth's name in class. Apparently they were in gradschool at more or less the same time (this checks out - Purdom was a 1966 PhD from Cal Tech, Knuth a 1963 PhD. The Wiki page on Knuth shows that he was also a professor there from 1963-1968.), and Purdom took his first course in Algorithms analysis from Knuth.

You really can't do better than that, and I don't blame Purdom a whit for dropping Knuth's name. Knuth is considered by many to be the father of Computer Science and is often credited with having more or less invented the field of Algorithms analysis. He's also celebrated as a founder of geek culture, and his quirks are legendary. (For example, Knuth has a standing promise with the world to pay $2.56 - a hexidecimal dollar - for every error someone finds in his books. Most of these checks go uncashed, however, as they are considered collector's items. He has written over $20,000 worth of such checks.) I don't know all that much about him myself. From what I do know, I know that he and I are very different people. This makes it possible for me to admire him.

People have their own reasons for admiring others, of course, but the way I understand it, admiration means that the person possesses characteristics you don't have but might like to have. So many people, when you ask them to list their heroes, list people they want to be associated with. It's a sly way of asking you to think of them as you think of that person. For example, if a person tells you they admire Dr. Martin Luther King Jr., it tends to mean that, rather than admiring his rhetorical style or his bravery (or plaigarism skills or womanizing, who knows?), they just want you to know that they suffer from white guilt and are taking affirmative action to resolve the problem. This isn't actually admiration as I understand it. Admiration, rather, means that someone is something you like but will never be. This is the case with me and Knuth.

From what I know about Knuth, he is methodical and self-disciplined to a zen monk extreme. I am exactly the opposite. I have no self discipline, and it's a constant problem in my life. I am only able to do or focus on things I'm interested in. This presents a huge hurdle to my ability to get schoolwork done on time, for example. (And in fact one of the main reasons I'm pursuing a career as an academic is because I don't think I can be assigned things to do for the rest of my life. It is crucial that I do things I want to do and find interesting - or else things don't get done. Not an attitude one can really have on a corporate payroll...) A cool Russian professor I had as an undergraduate said once that "life is a fight against laziness," and I took that to heart. I wouldn't call myself lazy, exactly. I do a lot more actual work than most people. But I would say that I'm undisciplined. And I do indeed take affirmative action to deal with this problem.

Today in Algorithms we had a midterm. It was much easier for me than I expected. I have been neglecting this class and was shooting for a 50 on the exam; I spent all yesterday cramming with that goal in mind. (Of course, this is a Computer Science course, so please understand that grades in that range, while bad, are not "terrible.") I think I will probably get a 75. But if I'm to get an A in the course, which I fully intend to do, I will have to buckle down.

One of the stories Dr. Purdom told about Knuth that seemed characteristic was that when he showed up at Case Western the dean gave the standard speech about how one in three wouldn't make it to graduation. Knuth took that seriously and so resolved to do all the problems in his Calculus book, whether or not they were actually assigned. Asked whether this didn't take up too much of his time, he responded that no, in fact it took up only slightly more time than doing only the assigned problems, the reason being that if you do all the problems you get really good at the subject and so can finish the assigned problems faster than you otherwise would. Or, in other words, "slow and steady wins the race." I have seen so many examples of this in my life - of people who have this kind of discipline getting where they want to go.

In that spirit, I have resolved to save my Algorithms grade by reading the first three volumes of Knuth's masterwork. Once a year I embark on a 6-week discipline project of some kind - usually vegetarianism. I've already filled that requirement this year (had no meat from the second week of January to the end of February), but it never hurts to do it twice. Finishing Knuth's book in 6 weeks will mean reading 50 pages a day - and I'll still come up 200 pages short, I think. But these are foundational textbooks in Computer Science; I should read them anyway, and this is a great excuse to do so.

Because blogging is something I enjoy, and because discipline is not my strong suit, I have created a second blog as a daily journal on these readings and to keep myself focused. I plan to delete it once I've finished the books - but in the meantime, I'll post links to it on The Only Winning Move. Content, of course, will be my thoughts on what I've read.

I wouldn't be Knuth for the world. I like my flaky personality. Stodgy people are good for the economy, which is great because the fact that they're around means I don't have to be one. But it's interesting to try to be one for a day (or six weeks, as the case may be). I can't really imagine what life would be like if I actually did everything I set out to do (and knew how to only set out to do things I would do - which is the real secret to being one of these people), but projects like this are a good way to get a taste. Wish me luck.