The Quest of the Algorithm of the Chimes of the Bells of the Clock of the Long Now

I’m on a bit of a quest at the moment, which is turning up all sorts of interesting bits and bobs (and distracting me nicely from the work I should be doing today).

Yesterday I listened for the first time to Danny Hillis‘ 2004 Long Now seminar on “Progress on the 10,000 Year Clock” (listen here). Like most of these seminars (which I really should write about in more detail some time) it’s well worth a listen, bristling with neato facts, insights, and mind-expanding ideas. (A less time-consuming but still usefully detailed introduction to the clock project can be found here.) One of the things which particularly caught my imagination was the discussion of the bells. Apparently His Enoship noticed that the number of days in 10,000 years is almost exactly equal to the number of permutations of ten things (365 x10000 = 3650000; 10! = 3628800) – so the idea is that once per day these ten bells will chime in a certain order, never heard before, never to be heard again. (Eno has released a CD of “bell studies”, which is on its way to me even as I type.) Hillis invented an algorithm for the ordering of the chimes, so we know what ordering will be played for any day in the next 10,000 years (Eno’s CD plays the sequences for January 7003) and as you’d expect, people have run with this idea. I have an itch to run with it too, probably (predictably) in Haskell.

So where’s the algorithm? I haven’t found it yet, to my great surprise. :) I would have expected it to be fairly easily available, but apparently not. It might be in Stewart Brand’s book about the clock (also on its way to me), but I sure don’t see it anywhere online. However, there are a couple of implementations floating around, so perhaps some reverse engineering is in order…

Sean Burke has created a web page for exploring the bell patterns, with visualisations and MIDI downloads – code in Perl (and a Postscript (!) version here). Not fancying reverse-engineering the Perl too much, I wrote and asked Sean if there was a better source for the algorithm. Apparently Danny’s original version is in Mathematica. I haven’t found it, but Sean says he’ll send me what he can find in a few days. Otherwise, I guess I’ll keep digging, ask around on the Long Now forums, etc.

Sean pointed out that, when reverse engineering, “there’s understanding, and there’s understanding”, and pointed me at this fantastic war story, which sounds like a Daily WTF candidate to me. I’ve been there in the past, and in my case it was spaghetti Perl I was banging my head against – not pleasant. Still, the whole process of reverse engineering, of picking apart the code slowly, gradually and gently teasing the tangled knot open, can be a wonderful thing in itself. Or it might just turn out that the Mathematica code is clean and easily Haskellised. I doubt it, though, from what I’ve seen of Mathematica. :-) I expect it’ll live in a much lower-level domain than I want to work in, which is, of course, more than half the fun. If I can take an esoteric algorithm in a difficult language and translate it into beautiful and readable higher-order code, that’d be something worth writing about. So, watch this space (but don’t, of course, hold your breath).

In the mean time, as I said, the search is turning up all sorts of cool stuff. We present:

Another version of the algorithm, by Joe McMahon, in OS X/AppleScript discussed here, here & here (code via that last link). Interesting mention of ChucK too, which looks quite shiny (though again, maybe a little low-level for my taste).

Prototype chime generator diagram – I would wear a t-shirt with that on it.

An interview with Alan Kay from two weeks ago, which also points at the Kay-says-it’s-a-must-read Doug Engelbart essays (and no, I must confess, I hadn’t heard of Engelbart).

Pop culture lives in the present; it doesn’t really live in the future or want to know about great ideas from the past. I’m saying there’s a lot of useful knowledge and wisdom out there for anybody who is curious, and who takes the time to do something other than just executing on some current plan. Cicero said, “Who knows only his own generation remains always a child.” People who live in the present often wind up exploiting the present to an extent that it starts removing the possibility of having a future.

Stewart Brand meets the Cybernetic Counterculture – whee, the 60s!

(These last two via this page on “admirable people”.)