Customer Reviews

Things a Computer Scientist Rarely Talks About (Center for the Study of Language and Information - Lecture Notes)

5 star:		(4)
4 star:		(6)
3 star:		(0)
2 star:		(1)
1 star:		(1)

 

 

About two-thirds of this book concerns Knuth's experiences writing a book entitled "3:16." Knuth decided to study the Bible through a sort of stratified random sample: taking Chapter 3, verse 16 from each book of the Bible and studying it in depth. These discussions do have their interest, but I do not feel that "Things A Computer Science Rarely Talks About" really stands on its own.

A better title for this book would have been something like "3:16: The Story Behind The Book."

Consider chapter 4. This is mostly a series of stories about how Knuth and Herrmann Zapf asked many of the world's leading calligraphers to illuminate the verses he used. He talks about the suble colors used to print them, the delicacy of the originals (not fully picked up by a 600 dpi scanner and requiring pixel-level editing to prepare them for reproduction). It therefore can only be described as irritating to be presented with these pictures, about fifty of them, in the form of little 2x3" halftone reproductions.

I do not feel the book delivers much on the promise that Knuth will reveal "the many insights that [he] gained" from the work. He talks a good deal about the reasoning and the process behind the project and it is quite interesting, perhaps even inspiring in the sense of making one wish to do likewise. But his presentation of his own beliefs is rather muted and low-key. I perceive this as modesty, not evasion. Still, it is not what I expected. Consider chapter 4, again: we learn a great deal about how he feels about the esthetics of the calligraphy, how he edited them, how he dealt with issues like calligraphers who inadvertently made mistakes in the text. But are there really any religious insights here? Well, subtle ones, perhaps.

I think he is sincere when he says, referring to "3:16," "I am not here today to sell copies of the book." In a discussion, someone asks "What would you recommend for computer science students who have never read the Bible?" and I believe Knuth is joking when he says "The number one recommendation is that they should certainly read my book. You know, it makes a wonderful Christmas gift. More seriously..."

Chapters 5 ("Glimpses of God") and 6 ("God And Computer Science") are fascinating, and come close to delivering on the promise of the book. I would gladly read a book-length expansion of this material in these two chapters.

Still, on finishing this book, I am aware of two feelings: a) an interest in reading "3:16," and b) an irritation with myself for having purchased this one.

In modern, Western, pluralistic, secular culture, it is currently still difficult for many to understand the interconnectedness of "science" and what is often called "religion", and to discuss many of the topics included in the text of this Knuth work. However, while this book could have been titled "Things that Society Rarely Talks About", the text specifically centers around the computer scientist and the software engineer. While the majority of "Things a Computer Scientist Rarely Talks About" is in essence a transcript of Knuth's lectures at MIT, a transcript of a panel discussion also included in the concluding pages of the book, as well as the Q & A sessions following each of the 6 lectures, provide societal context in that the thoughts of other thinkers in this field are given argueably equal treatment alongside that of Knuth. I would argue that the content of the fourth lecture is a bit too much centered around the author's work on "3:16", but I found this material fascinating, especially in light of Knuth's third lecture on the difficulties of natural language translation and how this difficulty is not as prominent in computer science. As a software engineer, I consider this book to be an extensive (although sometimes incoherent) expansion to the occasional spiritual comments made by Frederick Brooks in his classic software engineering work, "The Mythical Man Month" - more specifially, the pages Brooks wrote on the "joy of the craft" in which he compares the joy of the software development process to "God's delight in making things", and his comments on "the delight of working in such a tractable medium" where one is "only slightly removed from pure thought-stuff". Guy Steele, of Sun Microsystems, notes in the panel discussion that "...the study of computer science, as we know it, is a human activity, so far; human activity is spiritual, at least in part; and therefore the study of computer science is spiritual, at least in part. This may seem to beg the question. My point is that by human activity I mean to imply activity with purpose. I want to talk here about purpose. Why do computer science? Because it's fun? Because it is beautiful? To earn enough money to buy mango ice cream at Toscanini's [an ice cream parlor near MIT]? To improve the lot of humankind? To serve God? For me, it is ALL of the above." Knuth's lectures, the Q & A, and the panel discussion are all presented in what appears to be a sincere discussion of related topics that many would rather not discuss. Both Knuth and the members of the panel admit that they do not have all the answers, and that there is some confusion. For example, a self-proclaimed "devout agnostic", Mitch Kapor, who designed Lotus 1-2-3, here discusses that during the 1970s he became a teacher of transcendental meditation - of which he describes it as a "...very interesting experience. I can tell you that people really don't levitate. In fact, it is a cult, at least if you get highly involved with it. I have a lot of bruises acquired along the way from some of my involvement. I am kind of a Buddhist fellow traveler, in the sense that I have a lot of affinity for a Buddhist's way of thinking; but I just can't seem to make it in any organized religion, including Buddhism". So while Knuth is grounded in the Christian worldview, and is not a relativist, he as well as everyone else participating in this discussion has questions that have no certain answers - this fact must of course be one of the underlying premises of this book.

`Things a Computer Scientist Rarely Talks About' by the distinguished computer scientist, Donald E. Knuth, professor emeritus at Stanford, is a collection of six lectures given at MIT as a part of the "God and Computers" project started in 1997 by Anne Foerst, a faculty member at the MIT Artificial Intelligence lab.

For those who are not familiar with Donald Knuth's name, I can approximate his renown within his speciality as being comparable in stature to Bob Dylan in contemporary folk music or Julia Child in the world of American culinary writing. His professional accomplishments include the foremost texts on the analysis of computer algorithms and a groundbreaking creation of software for computerized typesetting. He is also a lifelong devout Lutheran who has written a prominent synthesis of biblical texts and calligraphy, entitled `3:16'. Much of this book takes us behind the scenes of how this book came to be. I will quickly warn you that as insightful as Knuth's lectures are, the introduction by Anne Foerst and the symposium at the end are not of comparable quality.

This volume is a great affirmation of how a person can not only embrace scientific and logical disciplines and still be a devout Christian. I am really happy to find Knuth especially beginning his fourth lecture with a reference to that great forgotten American philosopher, Charles Saunders Peirce (the inventor of Pragmatism) and how Peirce divided `normative sciences' into the rational, the ethical, and the aesthetic. Not only does this show that matters of religion, closely allied with the aesthetic, are really a different world of discourse than either logic or questions of right and wrong. (The distinction between morality and religion is an important matter, but not an issue in this book.) The main thrust of this distinction explains why scientific or mathematical discourse is actually not directly relevant to matters of faith, just as matters of faith are not directly relevant to issues of scientific truth and its theories.

Logic and Science have locked horns with faith for centuries with all sorts of mistaken concepts such as the proofs of the existence of God done by the Scholastics and the Great Clockmaker of Newton's contemporaries have fallen by the wayside. The first is a mistaken attempt of church scholars' shoring up faith by reason. The second is the mistaken attempt of scientists to rationalize God in the mechanistic world of Newtonian physics. Knuth skirts all of these misguided efforts and provides a wealth of new perspectives on how faith in God can live comfortably with a total embrace of modern physical science, mathematics, and logic.

The first lecture introduces Knuth's ideas behind his 3:16 project that began as a discussion group at his church on the 59 verse 16s in the third chapter of books in the Protestant bible. The choice of 3:16 is almost random. The primary determinant was that he wanted to be sure that there was at least one important verse among those discussed, and John 3:16 is about as important as you can get to Christian doctrine.

The second lecture reveals many important and interesting aspects of random numbers in modern mathematics and computer science. For the non-computer theorist, surprisingly, a random sampling algorithm often arrives at a solution to a problem much faster than a brute force examination of all possibilities. This liberates us from the mechanistic worldview and shows how chance and its important human facility, free will are a lot closer than one may think. And, Christianity simply would not make sense without the reality of free will!

The third lecture deals with some of the many problems arising in the translation of biblical texts. The most immediate lesson I take from this is to reinforce my skepticism of all Christian groups who put an uncritical faith in literal Biblical statements. Knuth reflects on several of his featured verses where a difference in translation literally changes the meaning of the text in English. This explains why Knuth says that his translating these verses for himself was one of the very smartest things he ever did.

The fourth lecture deals with aesthetics and the importance of beauty in the effective presentation of ideas, especially ideas from religious texts such as the Bible. His source for this chapter is the stories behind the creation of his 59 pieces of calligraphic art commissioned to illustrate the 59 verses. The artistic rendition of each verse was done by a famous and talented calligrapher (Note that one of Knuth's two major accomplishments is in the computerization of fonts and typesetting.).

The fifth lecture states what it is about God that Knuth learned from his 3:16 project. I will attempt to summarize Knuth's findings with two statements. The first is that the payoff of faith is not the `eternal life' promised at the end of our physical existence, but the active search for the eternal in our lives on earth. The second is that the primary objective of the search is to do their best to be in harmony with God's wishes.' The fact that the second objective requires a lifelong search is more than adequately explained by the fuzziness in interpretation of the not entirely perfect translations from the Aramaic and Greek texts with conflicting and imperfect sources.

The sixth lecture deals with God and computer science, as well as God and other sciences such as quantum physics. While the primary subject of earlier logical tools was based on randomization, this lecture was based on the effect of very, very, very large numbers and our concepts of the universe, and how that all fits with our notions of God.

I'm particularly tickled by Knuth's ending with a reference to Leibniz' notion that we are living in the best of all possible worlds, as Herr Leibniz was so easily parodied (by Voltaire, for example) that his good point has become a comic book caption.

This is an eye-opening book that celebrates Christian faith while making no compromises with logic.

First of all, this book is a transcript of the lecture series that Knuth gave in MIT in fall 1999. The nature of the content has to be taken into account when reading the book, because there are constraints in lectures that limit the amount and depth of what can be said, compared to that in an actual written book.

A significant portion of the lectures (the first 4 out of 6) were devoted to the 3:16 project. It is his effort to get a random sample of the verses in the Bible and study them deeply. It is understandable that Knuth did this, because, as he said, first, he admitted that he doesn't have formal training in religious studies, so he had to speak from his own personal experiences, and second, that the project was the turning point for him in his view of religion and God. Granted, you might be disappointed if you're looking for the ultimate explanation of how computer science and God relate to each other. The last two lectures dealt a bit in finding this explanation, but it seems that because of the constraints of time and scope for the lectures, Knuth didn't dwell too deeply on this issue. Overall, the book is satisfactory as Knuth's personal account, as a computer scientist, on his relationship with God.

If you are a mathematician or computer scientist who has an interest in the Bible or religion, this is stimulating reading from an unexpected source. If you are a mathematician or computer scientist with NO interest in the Bible or religion, skip this. For myself, after reading the other reviews, I found the discussion of Knuth's 3:16 project interesting and not a commercial. (you could work it the other way too: maybe if you buy this book, you DON'T have to buy 3:16). I admit to being a little disappointed by Lecture 6, but after the build-up, I'm not sure anything would have lived up to my expectations. Recommended (as long as you observe my caveat above.)

Knuth, known for his work in computer science, uses this book to talk about how religion and science don't necessarily have to go against each other. Although an easy read, it does drag at points whenever he talks about his work on the 3:16 project. Good for a weekend leisure book.

Most of the book is about his other book, 3:16. Only near the end do we get to read what Knuth thinks about God, and here, you will be very disappointed. He trots out a few mundane ideas, like that finite numbers, while not infinite, can still be very big, and therefore adequate for describing the universe and God. Another (unoriginal) idea is that perhaps the universe is one big program similar to Conway's Game of Life (cellular automata - also now discovered/invented by Wolfram), with some randomness thrown in for free will. Finally, there is a panel discussion with a few others (Guy Steele, Manuela Veloso, Harry Lewis, Mitch Kapor). What an incoherent mess! At first, I felt sorry for these people, as they are dealing with a difficult topic, and it is hard to be coherent in real time. They all basically try to use computer science metaphors to describe religious concepts, and do a fairly poor job. That's understandable. But they are also quite annoying, as they are all a pretty self-centered bunch, constantly quoting themselves or work they've done (that doesn't advance the topic at all), and ultimately, you get the feeling they each think they ARE God, with all the "great ideas" and programs and robots they've created and imparted with their wisdom.

If you are a computer scientist, you will be disappointed. If you are a theologian, please don't think Knuth (and the others in this book) are representative of other computer scientists as far as this topic goes.

... Definitely a waste of $....

As other reviewers have pointed out, this is a book about a book (3:16). The last chapter and the panel discussion form the meat of the book. What is more, existence of God is an axiom, and words like "God' will", "God's children" are tossed around frequently. If you are uncomfortable with that and have come to this book hoping that a world reknowned computer scientist would share his thoughts on these fundamental questions (such as whether God exists and if He does, does he have a plan for us etc.), you would be disappointed.

As a computer scientist, Donald Knuth is not expected to be a story teller. And he is not. But Donald Knuth is a terrific fact based writer. He knows how to think deeply, how to get deeper into a subject, its key elements, and their synthesis overall. Moreover, he knows how to write and present that to others after his thinking gets over. So, when God and Computers are to be brought together, he has left his mark through the six lectures he delivered at MIT by invitation.

The lectures, however, draws a lot of examples from his project on Bible 3:16. At times, it seemed to be better to read his book on Bible 3:16 first, and then read this lecture series. He has been candid in stating his belonging to a specific branch of the Christianity (Lutheran), and at times the lectures gave a feeling that a thorough study of Bible (or history of Bible) would be helpful before reading this lecture series.

Setting that apart, there is enough food for thought to a computer scientist. The first lecture states when it comes to theology Mr. Knuth is a user, not a developer. That instills a unique feeling to every programmer and computer scientist I am sure: when it comes to programming this tribe sits in the developers seat and they can now imagine how they can be perceived by the users if the work is done right.

My second takeaway was the statement that talked about the importance of rational expression of all emotional feelings, but how hard it is to do in reality.

The third and final takeaway was his experience in writing "Surreal Numbers". The story of how he spent six days in Oslo to write it up, and how his mind was blanked out on the seventh day, was simply amazing. The analogy of the muse sitting on his shoulder for six days and then moved off was quite profound.

A must-read book for those who think themselves as computer scientists.

What trait does Knuth think he has in common with Bill Clinton? Are the assumptions of Chaos Theorists too real? Does Godel's Incompleteness trump Heisenberg's Uncertainty in discussions of faith? These and other nerdy nuggets are skillfully woven into Knuth's Alfred E. Neumann-meets-Albert Einstein excursion into theology.

As with Knuth's technical writing, I was lulled into a false sense of complacency by Knuth's colloquial style and sense of humor. While this book is a transcription of lectures as well as Question & Answer sessions given by Knuth @ MIT discussing his modus operandi for writing _3:16 Bible Texts Illuminated_ , it was a much more thought provoking complement to _3:16_ than I had expected. For example, framing the discussion of man's Free Will and God's Omnipotence from the perspective of the Complexity Theory (from the Father of CT no less) sheds fresh light on this age old debate for theologians and philosophers alike.

Knuth has the gift of being a serious scientist without taking himself too seriously and in the end, scientists may discover that Douglas Adams is right and that the Answer to the Ultimate Question really is 42. Yet, with this book, Knuth beckons us to enjoy the ride. And to keep asking our own questions.