I could read this every year for quite a while

I'm on page 489 of my second reading of Anathem, which I've been putting off for months now. I read it over a year ago, and there are bits of it that spoke to me a lot -- the way the main character just barely is accepted by the Edharian order but really wants to be there, the deeply dorkalicious ways of most of the characters in the book, the linguistic playfulness and send-up of our current culture. Reading it a second time is just making me happy, and more sure than ever that in general it is the right thing for me to do technical work, study technical things and hang out mostly with people who are really into what they do. It feels good to be so sure -- and it feels strange that a work of fiction should be the thing that can make me so sure about real life, although I think that happens all the time.

(I also want again to work my way through either my old calculus book or a beginning probability book, not in a huge hurry but for fun. Could this be a reasonable project for Christmas vacation, or will I still have enough left of my software side project left that I'll feel obligated to work on that? We'll see. I'm in a back-to-basics intellectual mood, lately, and less interested in bulshytt.)

It's hard not to daydream about some kind of acrobatics-and-math-and-software concent, with some music for heisenbug and only a small number of personal possessions and lots of cats. But then, my daydreams are probably just me asking myself permission to live that way.

As an example

I don't post terribly often about technical things here. Blogs are are so useful for venting emotion that I often use it that way, and anyway I think people probably like the acro pictures better. Rest easy, I'll post some of those next.

Still, I do think about math and logic and stuff, especially when I'm falling asleep and need to think about something to calm my mind down. (It's nice, though it also slows my thinking a lot, since I tend to fall asleep before getting anywhere). The last few nights it's been this puzzle, more or less:

You have n bits in a row. Each bit can be either 0 or 1, but you can't have two 0s in a row. What is the expression of how many patterns you can have in n bits, and why?

I knew the pattern all along because someone kind of gave it away, but it's pretty easy to work out like this:

One bit -- valid patterns are 0 and 1 -- two patterns.
Two bits -- valid patterns are 11, 01, 10 -- three patterns.
Three bits -- valid patterns are 111, 011, 101, 110, 010 -- five patterns.
Four bits -- 1111, 0111, 1011, 1101, 1110, 1010, 0101, 0110 -- eight patterns.

The sequence 2, 3, 5, 8 should probably be familiar if you inherited any math-dork genes, and it's not an accident; that pattern does continue. The next one is 13, then 21. Each number is the sum of the two numbers before it.

Now, can you prove or explain why the valid bit sequences in the puzzle would form a sequence like that?

1) Comments obviously might contain spoilers.
2) If you don't like bits, you can think about a staircase where you can step on (or bounce a ball on) every step or every other step as you traverse the stairs.

And now you can have pictures

From acro camp a few weeks ago:
( Three pictures )

From last Tuesday night, my first successful (if short-lived) loop-de-loop with Amy: