Douglas Hofstadter

Faculty page at Indiana University: http://www.cogs.indiana.edu/people/homepages/hofstadter.html

I always remember DouglasHofstadter for his column in the ScientificAmerican - Mathematical Recreations. As I remember it, the columns often focused on the concept of "self-reference" and recursion. See ThisIsTheTitleOfThisStory as an example. (MetaMagicalThemas is a collection of these articles grouped into seven themes, two of which relate to recursion. -- FalkBruegmann)

The column was actually called Metamagical Themas - just like the book. His predecessor, Martin Gardner, ran the column as Mathematical Games, of which Hofstadter's title is, of course, an anagram. -- DavidHarvey

I also remember his participation in CSICOP ("psi-cop"). That's the group that mocks Coco the Gorilla as a fraud that only does sign language when her trainer is nearby. -- PhlIp Actually, I believe that Koko has been videotaped signing at night to herself when no one is around. I don't know sign language, but I think videotapes of Koko are available for the curious; I bet transcripts are also available, although you probably have to email the researchers involved. Koko has taken a human IQ test and scored reasonably well (>75?). -- BayleShanks

He also wrote Gödel, Escher, Bach: an eternal golden braid. I've never had time to read it, but people have told me they liked it. -- ShalomReich

He also wrote TheMindsIbook, with others.

Introduced PeterSuber's NomicGame to a wider audience.


One of the things I liked the most about GoedelEscherBach was the use of a Platonic dialogue between two individuals (achilles and the tortoise) as an illustration of the topics that he would cover in more conventional prose later in the chapter.

This kind of dialogue used to be a common theme in scientific writing up to the time of the enlightenment, when it seems to have fallen out of favor. For instance, Galileo used it in a book explaining the Copernican model and explaining why it was superior to the Ptolemy's model of the universe.

In the book that BobbyWoolf, ShermanAlpert & I are writing, The DesignPatternsSmalltalkCompanion, I use this kind of dialogue to introduce various ways in which design patterns are used. We then cover them in the text in more rigor and detail. I will freely admit that my inspiration came from Hofstadter, and in fact I read GoedelEscherBach again just before writing it.

-- KyleBrown

Comment on Kyle's digression: Dialogs in G? work just great, but I remember how infuriating they are in Plato! There is one stupid guy aside who gives ridiculous and often slavish comments, and systematically misses any interesting question or counter-argument! -- Marc Girod
I found his newest offering, LeTonBeauDeMarot, a fabulous, world-view-shattering work, even more influential than GoedelEscherBach. -- TomStambaugh

LeTonBeauDeMarot offers deep insight into how self-imposed constraints are the key to tapping your creativity. I found that reading the book, and as he suggests, making my own translations of the eponymous poem as I was influenced by the translations in the book, trying out various constraints and limitations, turned the experience into something very private, and very universal at the same time. -- MathieuGlachant


I found GoedelEscherBach interesting, but too much of the same to be able to read the whole thing. I have read the first half twice, and only selected parts of the second half. The main theme seems to be GeeWhizIsntRecursionGood?, but once you understand why FixedPointCombinators work, recursion only holds so much attraction. The stories about Escher and Bach are interesting, and ZenKoans are always an inspiration. -- JohnFarrell

It's an age since I read that stuff, but I always figured he picked the wrong composer and that, say, AlbanBerg? would have been more appropriate. Berg also loved to play mathematical games in the score, although they're harder to hear than in Bach, and the complexity of his view of the world seems a better match than Bach's solid faith. -- SteveFreeman

Yeah, the Bach parts of GEB are really bogus and completely miss the point of Bach's music, only some of which is recursive and only in the faintest way. The really interesting thing about that music is not that it has sophisticated internal structural relationships but that Bach was fluent enough in those structures to be able to write musical music despite them. -- D. Collinge

Bach's solid faith may be unfashionable these days, but that is no reason to take digs at it.

see DRH's contribution to the field of project planning: HofstadtersLaw.


DRH is brilliant.


Does anyone know if RichardStallman is a Hofstadter fan? I have always assumed that the inspiration for the name 'GNU' (GNU Not UNIX) came from GoedelEscherBach, but I could be wrong. there is an entity in GEB which is a nested hierarchy of djinns (rather like TheCatInTheHat?), where all the djinns above the current one are referred to as GOD, standing for GOD Over Djinns. -- DaveKirby

The use of recursive acronyms by hackers predates GEB. For instance, EINE ("EINE Is Not Emacs") and ZWEI ("ZWEI Was EINE Initially") date to 1978 and 1979 respectively; GEB was first published in 1979. RMS was inspired by the hacker tradition, not by GEB. I don't know whether the GOD acronym in GEB was inspired by hackish recursive acronyms. -- GarethMcCaughan


The first time I read GEBTEGB I also took away the impression that the book's main theme was recursion. Subsequent readings have revealed, to me anyway, that it's about complexity and its application to AI. Recursion is just a handy tool that helps deal with complexity by introducing more of it.

One of my favorite New Bits of Knowledge was the observation about how humans instantly know to cut when presented with a recursive paradox, but we haven't a clue how to teach a machine to do it.

-- BobBockholt


CategoryAuthor

EditText of this page (last edited September 11, 2007) or FindPage with title or text search

Meatball   Why   Wikibase