User:Rspeer

I'm Rob Speer. I started this wiki, and if you're interested enough in reading this instead of other, more exciting pages, I'll describe what I'm hoping for it to be.

Explorers do better with a map
I've seen so many people who start reading the book, get excited by the ideas it's presenting... and then get left hanging, without any idea of how to learn more.

There are many directions one can explore starting from GEB. You can seek out the Bach music and listen to it. You can learn a functional programming language that's good for expanding on the CS theory that appears in the book. You can try to find out where AI research has gone in the last 35 years. You can follow up on the philosophy of consciousness (a topic where nobody, and certainly not Hofstadter, really has the last word).

But very often, you're doing it on your own, putting great effort into unknowingly re-tracing the steps of others who have read the book.

Some of these explorations are dead-ends. One criticism of GEB is the number of people who read it, or half-read it, and think this alone qualifies them as philosophers; then they start spouting the half-baked philosophy you would typically expect to hear from someone who is high.

It's possible to follow paths from GEB that are stepping stones into serious theoretical math. But if you follow the wrong paths, you may end up stepping into meaningless mysticism and numerology.

I don't think that's the fault of the actual content of the book. But it might be a danger of the form. The book takes you on an intellectual journey, and then it ends, leaving you far from home without a map.

So far, a lot of the context surrounding GEB is a sort of oral tradition. Oral traditions lose information. The discussions on GEB that I started at MIT and online made me realize this: We can help future readers of the book by taking these explorations and writing them down.

My connection to this book
I found GEB hugely inspiring when I first read it, and I want other people to be able to share the experience.

In 2006 and 2007, I organized seminars at MIT where we met once a week to discuss chapters of GEB. It was organized through a program called ESG where students can just teach each other things. People got a small amount of course credit for attending it, so it was in the course catalogue and everything.

One participant in the seminar, Justin Curry, was inspired to start his own course, together with Curran Kelleher. (See Curry and Kelleher.) This was in a different program, ESP, where college students teach high schoolers. The course was recorded on OpenCourseWare, so as a result, some nice lecture notes from the course are online.

Do keep in mind that despite that these were "MIT courses", in that they were courses taught inside MIT buildings, there were no professors involved in creating either one of them.

I graduated from MIT with an M.Eng. in computer science and a B.S. in music. (Yes, a Bachelor of Science in Music. I don't know, ask MIT.) You might see the influence of GEB there. Later, in 2012, a Reddit thread came across Justin and Curran's lecture notes, and I explained where they came from, and suddenly a few thousand Redditors were saying "YOU HAVE TO LEAD A REDDIT READ-THROUGH OF GEB".

I did. But I was also starting an AI company at the time, and I didn't keep these thousands of readers organized enough, so the read-through drifted apart in Part II. In 2014 I tried again, with no more success.

I concluded that Reddit is the wrong medium for this. It's made of distractions, discussions are forgotten over time, and it seems impossible to keep a discussion together when a large number of readers are reading at very different paces. The fast readers forget about the discussion by the time it finally happens, and the slow readers give up.

That's where this wiki comes in. I think that this is a medium where we can do it right, and create a shared experience of reading and understanding GEB.