Hacker News Reader

Comment by AllegedAlec | Hacker News Reader

Except that Goedel, Escher, Bach is not as pretentious...

I found Goedel, Escher, Bach just as pretentious and ridiculous as this paper. As an undergrad, I wrote a paper that was almost that pretentious, and I'm embarrassed about it every time I think of it. How do these things make it past, in GED's case, editors, and in this case, both referees and editors? Someone has to point out the Emperor's unfortunate nudity.

You're not wrong that GEB is grandiose and presumptuous in the material it takes on. But one important difference may be that LJA has started from one person's very particular choices* and tried to tie them up and cast them as universal. That often doesn't come off well: I think it tends to be more about the author than the reader. (Caveat: I say this only having skimmed the paper here, not read the whole thing. And having loved GEB.)

GEB, on the other hand, mostly started from a bunch of universal ideas, and used the author's idiosyncrasies as a way to explore those ideas. I think it comes much closer to hitting its mark.

There's no accounting for taste, though. If you don't like Hofstadter's way of thinking, you probably just won't enjoy his book. I'm sure some people find the Dialogues and their wide-eyed "discoveries" of analogies with the essays' material to be smarmy at best. Personally, I find the language of this paper to be overly flowery and too cooler-than-thou (yes, I've heard of Erik Satie; does it matter?) to enjoy.

----

* I mean, "By introspecting my personal aesthetic sensitivities, I eventually realized that my tastes in the scientific, artistic, and physical domains are all motivated by the same driving forces"...that basically says "I think you're going to love reading my diary".

Here is another piece of my personal experience with GEB. In GEB, I first encountered SHRDLU and the "blocks world". But, you see, I didn't think it was real; I somehow thought it was a use case document for such a system presented as a fantasy interactive session, haha! There are so many "fantasy fugue" type threads in GEB that you don't know what is real and what is fiction. Hofstadter doesn't go out of his way to make it clear. I don't remember seeing a clear clue that "hey, ignorant reader, this is a real session captured from a program someone actually wrote in Lisp in early 1970-something". At that time, there was no (residential) Internet and no search engine where you could just look up SHRDLU.

Shrug. I found that arithmoquinification of G's uncle stuff to be very helpful in terms of its explanatory power. Puffed up or not, Hofstadter made it accessible without dumbing it down. GED is a good overview of various issues and stimulates thinking.

He kind of lost me when in one of his books, Le Ton beau de Marot, I came across a little diatribe against rock and roll. Fuc...king wanker! :)

I certainly disagree that GEB was pretentious. For beings that can understand information, surely computation is the highest possible goal. To that end, a certain seriousness and prestige is available when referring to CS.

We all have our opinions.

Did you find GEB to be a helpful reference? What kind of science were you studying?

Not to invade your privacy, I was just curious.

> For beings that can understand information, surely computation is the highest possible goal.

That sounds almost like a statement of religion. I am a being that understands information; "computation" is not even remotely close to my highest goals.

Maybe not as an individual, but as a society or as a species, there are few absolutes for us to pursue. History and Math are the only absolutes we really have.

What I'm meaning to say is its a righteous goal compared to pursuing art or literature, because it's universal in nature.

I think Turing will be remembered as one of the few people who changed the course of human thought in its entirety.

I understand what you're saying, and sure, my goal right now this weekend is to get some plane tickets and an airbnb booked, there's no computing mysticism there.

But as a whole, I think we humans have an objective to find four the most information possible and to do the most computing as possible. That's essentially what the singularity is about, a human creation coming to a point of autonomy. That's the next step for humans, I think, abandoning flesh and spreading autonomous machines throughout the cosmos.

I guess it IS my religion.

I didn't find GEB helpful. It presented its information in a way that, if you're being charitable, sounded like, "Check out this cool thing these interesting people have in common that will tweak your brain!" The problem for me was that it often lapsed into, "Check out how cool I am that I can understand these acknowledged deep thinkers, and impose this pattern on their thought even when it's only peripherally appropriate!". Some of that perception may have been from my assessment of the people I knew who talked about it too much, though, now that I think of it.

The language in GEB can be a bit much, but I find it to be a consistently effective took for opening someone's eyes to the vast universe of computer science.

There should be a GEB for Hackers or something like that, though. Accessibility is an issue. Maybe not accessibility in the sense of grokking the content, but certainly in staying engaged to it throughout the text. It feels like a lot of work.

Humility is important too. GEB could definitely use some more of that too.

Thanks for your insight

GEB touches on the internal looping of G, E and B. There is a real idea underneath.

Pulling out my curmudgeon hat for the sake of argument... I read GEB in college and actually bought a copy about 10 years later, to see if it resonated more. But I ended up barely re-reading it.

I don't remember anything that substantive about the connections. "Internal looping" seems like a vague connection.

The Godel sentence is self-referential for sure. That's a concrete and specific property.

All I remember about Escher is that there is the picture of the hands drawing each other. OK, it's a great picture, and a nice analogy. Is it profound? Did I miss something besides that trivial observation?

I don't recall how Bach is supposed to figure in. How is Bach self-referential or how does it display "internal looping"? (And what does that even mean?)

Computer science is full of better examples than Escher or Bach, IMO. For example, C compilers written in C, which lead to the Thompson "trusting trust" attack. Compilers are mentioned in GEB, but it's done in a way that I didn't find particularly helpful (probably because I already know something about the subject.)

I also experienced this with my Python compiler written in Python [1]. There were some interesting cases where knowledge of the variable names used in the compiler "leaked" into the compiled output via, I believe via the string intern table of the Python VM.

[1] http://www.oilshell.org/blog/2018/03/04.html

Escher paintings are circular and self referential. Staircases that perpetually go up.

Bach’s music is circular in terms of scales. A up to G and back to A.

But circular != self-referential. Self-referential is a deeper property.

If I have 2 mutually recursive functions, that's circular. For example, most parsers have mutually recursive functions.

There's nothing self-referential about them though. On the other hand, a C compiler written in C is self-referential. It confuses subject and object, which leads to some interesting paradoxes like the number for the "tab" character never appearing in the source code, as Ken Thompson points out.

The Godel sentence is self-referential. It makes a claim about itself.

The movie Synecdoche NY and some of Lewis Carroll's works (which figure heavily in GEB as I recall) are self-referential. The person writing the work appears in the work. Charlie Kaufman does this a lot -- the movie Adaption also has these properties.

But I don't see that Bach's music is self-referential. The music doesn't write more music, or have any awareness of itself.

Escher's two hands is self-referential, because it's confusing subject and object. But the staircases aren't self-referential.

G.E.B. did win the 1980 Pulitzer Prize winner for a work of general for non-fiction.

But Pulitzer Prize committee likely don't have any experts in math, and can't verify some bold statements about math theory described in the book.

Oh, well, if it won an award, it must be good. /s

Pretentious just simple extension to CLOS (Common Lisp Object System) thanks to the power of Lisp and MOP (Meta Object Protocol).

In any other language adding aspect oriented, context oriented or pretentious extensions requires writing a whole new language.

Reply on news.ycombinator.com