I'm not cut out for this. Every now and then I think it'd be nice to write a paragraph every day, but then when it comes to the end of the day, I say "eh, I can write about that tomorrow", and by the time I get around to it I've forgotten what's happened.
This weekend was the annual local juggling festival. We got some coverage from the Ann Arbor News this year, but attendance seemed about the same as usual. For me the most fun coverage is from Homeless Dave, who brought his teeter totter for us to play with in the morning and did a few interviews.
I ran the games this year. I didn't have the time to prepare the week before, so it was a bit rough, but people seemed to have fun.
Sara and I went out to dinner with a group to a Korean restaurant named Seoul Garden. I had some bibimbop. I felt like they skimped a little on the side dishes (always my favorite part), but it was otherwise great.
Then to complete the already long day there was a party at Fred and Deanna's.
They had a puzzle in the form of a dodecahedron with a lit-up button on each face. It starts with a random configuration of lit and unlit buttons. Pressing a button toggles the lights on the five neighboring buttons. The goal is to light all the buttons.
My math is getting sadly rusty, but I still recognize when something is linear, hence easy, so I fooled around a little, and eventually realized the nice way to look at it: given a button, call it "odd" if an odd number of its five neighbors are lit, and "even" if an even number of them are lit. In the winning configuration, for example, all twelve buttons are odd, since all 5 (an odd number) of each button's neighbors are lit up.
Then the trick is to notice that pushing a button changes the parity of the button you pushed (because if n of the 5 neighbors were lit before, 5-n are lit now, and n is odd iff 5-n is even), but doesn't change the parity of any others (it's easy to check that every other button borders on either zero or two of the buttons that are being toggled, and that toggling two neighbors can't change a button's parity). It follows that you can produce any desired pattern of odd and even buttons. And that there's a very simple algorithm which always produces the shortest possible solution: just find an even button, press it, and keep going until no buttons are even.
Anyway. Today we were pretty tired out. We both napped some, and I did a grocery run in the afternoon.
Wednesday I leave with Fred and David for Connectathon. I've been using kvm to set up virtual machines on my laptop to use for testing. It's been working really well. Effectively it means I'll have all the test machines I could want without having to carry more than the one laptop.