dive into mark

The other day, as I was skimming Reddit for my daily fix of mental junk food, I stumbled across an old lecture by Douglas R. Hofstadter, Analogy as the Core of Cognition. A bit of URL-hacking then led me to this introduction of the man himself, which briefly describes his major works, including Gödel, Escher, Bach: An Eternal Golden Braid, which I first read in a delightful Logical Systems class in college, and which will be the only book in my backpack when I get to step 7 in my pursuit of happiness, though by then we need to be able to buy unencumbered PDFs of such things, because damn — and I mean this quite literally — that book is heavy. Seriously, Amazon pegs it at 832 pages and 2.4 pounds. That’s almost enough reason to buy a Kindle. Almost.

Hofstadter also wrote a book called Metamagical Themas, which before that was the title of a column he wrote for Scientific American in the early 80s. His column was the successor to Martin Gardner’s Mathematical Games column, which later served as the inspiration for several of my early Free Software games. “Metamagical Themas” is, of course, an anagram of “Mathematical Games,” a fact which I somehow missed until it was pointed out to me by a now-ex-girlfriend for whom Hofstadter was also a literary hero.

Eventually, the introduction led me to this page of “Hofstadteriana”, which includes Autoportrait with Constraint (written without the letter E — a feat I have also attempted), a selection of ambigrams (like optical illusions, but with words), and this potpourri of other random excerpts. This essay about what GEB:EGB is really all about is interesting, too.

(Good writing, bad writing, great writing, and the differences between them, is something I’ve discussed here before. But the letter to the editor described in For Meta, For Verse just makes me want to throw away everything I’ve ever written and go live in a wooden hut. Let me put it another way. On a good day, Techcrunch is bad writing. On a bad day, I strive for at least good writing. But what Douglas does is… something else altogether. That poem is proof.)

After reading about ¾ of the introduction page, I noticed that the first word of every paragraph was bolded, and furthermore that the last letter of every paragraph was italicized. Aha! The game is afoot! My suspicions were confirmed shortly thereafter. The initial bolded letters of each paragraph spell out “D-O-U-G-L-A-S-R-H-O-F-S-T-A-D-T-E-R,” which spells “Douglas R. Hofstadter.” But the final italicized letters of each paragraph seemed to spell out nothing at all: “l-a-s-i-p-a-n-y-e-h-t-r-o-w-n-e-l-g.” But of course! If the first letters spell something forwards, the final letters should spell something backwards. And so they do: they spell “Glen Worthey napisal.” Checking the page footer confirmed that the entire text was indeed written by one Glen Worthey, and as best as I can tell, “napisal” is a Polish word meaning “author” or “authored.” Glen Worthey wrote this. Aha!

Meanwhile, a bit closer to home, I recently learned that the appendices of the CSS 2.1 specification are in alphabetical order. I’m going to repeat that because it’s doubly awesome: the appendices of the CSS specification are in alphabetical order. “Appendix A. Aural style sheets,” “Appendix B. Bibliography,” “Appendix C. Changes.” Hmm, could be a coincidence. “Appendix D. Default style sheet for HTML 4.” Suspicious, but plausible. “Appendix E. Elaborate description of Stacking Contexts.” OK, now you’re just fucking with me. And so on, up until “Appendix I. Index.” (But, even more interestingly, not including Appendix H (!), which is absent from both the “mini-TOC” and the full table of contents, but which is present within the text of the specification itself, for example by going to Appendix I and clicking “previous.”)

Early on, I had tried to do something similar while writing Dive Into Python 3. Unfortunately for fans of constrained writing everywhere (both of you), the constant rearranging, merging, adding, and removing chapters led me to abandon that effort. But learning about the easter egg in the CSS spec prompted me to try again, and now if you skim through the online edition, you’ll see that the initial drop caps in each chapter are in alphabetical order from A to U. Maybe it’s not Hofstadterian, but I’ll be damned if I’m going to be outdone by the appendices of the CSS specification. Still, if the pen is mightier than the sword, I feel like I’m still learning how to unsheath the damn thing and wave it around. And Douglas R. Hofstadter is a level 60 samurai.

§

Eleven comments here (latest comments)

1. Nicely self-referential.

   — kasei 

2. He gave this talk in September 2009 at Stanford University if I read this correctly:
   http://www.youtube.com/watch?v=n8m7lFQ3njk

   — Robin Millette 

3. Knuth did it first in The TeXbook and the METAFONTbook, each with ten such lettered appendices.

   — Josh Lee 

4. Correction, the lecture I posted was recorded in February 2006. It was dated 2009 on youtube. More http://www.google.com/buzz/pilgrim/RdnKnvQXUAr

   — Robin Millette 

5. Thanks Mark. I enjoyed the clinamen in Hofstadter’s autolipography.

   — Thomas Guest 

6. For what it’s worth, “napisal” is the simple past form in the male third person of the verb “to write” in Slovenian. Like saying “[he]wrote”. Funny thing, Slavic languages.

   — Carlos Yoder 

7. The game is afoot :D

   — Peter Stuifzand 

8. Ah-ha! Thanks for explaining the changes to DIP3. I hadn’t looked at them in depth, so I didn’t really understand why you modified the initial paragraph in every chapter. Trying to preserve the letter game will be a nice translation challenge. I just have one further question: who’s that Philip guy you mention in the Mercurial commit message?

   — Giulio Piancastelli 

9. The Governator did it as well!
   http://www.nytimes.com/2009/10/29/us/29arnold.html

   — BennyB 

10. Right, Knuth did it before. In case someone is interested:

    The TEXbook:

    A Answers to All the Exercises
    B Basic Control Sequences
    C Character Codes
    D Dirty Tricks
    E Example Formats
    F Font Tables
    G Generating Boxes from Formulas
    H Hyphenation
    I Index
    J Joining the TEX Community

    The METAFONTbook:

    A Answers to All the Exercises
    B Basic Operations
    C Character Codes
    D Dirty Tricks
    E Examples
    F Font Metric Information
    G Generic Font Files
    H Hardcopy Proofs
    I Index
    J Joining the TEX Community

    — Thomas Hühn 

11. I read your post here which turned out to be more interesting and wandering than the title indicated. Then I read *Analogy as the Core of Cognition* and Hofstadter’s ideas of chunking and analogies and how concepts are built upon earlier concepts, etc.

    Later in the day I stumbled upon [Triumph of the Cyborg Composer][1]. A composer named David Cope wrote software that created original music which was indistinguishable by music experts from pieces created by skilled human composers. (It passed the *Musical Turing Test* if there was such a thing.) The software was originally designed to mimic the style of a certain musician. (Bach in this case.)

    Cope has a theory that nothing is original. Musical compositions are created from mentally borrowing and rearranging things we’ve heard before and piecing them together in a skillfull manner. To me, this sounds a lot like Hofstadter’s thoughts on analogies.

    I’ve actually read about Cope in the past but it was nice to have these new connections so I kept reading.

    As you might have guessed by this point, Hofstadter is actually mentioned in the article. He knows of Cope and his work and has talked and written about it. He is not so happy, though. Turns out Hofstadter is a kind of hopeless romantic.

    A good read. You should check it out if you haven’t already.

    [1]: http://www.miller-mccune.com/culture-society/triumph-of-the-cyborg-composer-8507/

    — Matthew 

Respond privately

I am no longer accepting public comments on this post, but you can use this form to contact me privately. (Your message will not be published.)

§