March 11th, 2002

beartato phd

(no subject)

Didn't get to sleep until about 1:30 last night. Kept hacking
random front-end things onto the rule 110 code. In particular
it can detect big patches of background and color-code them
by phase. Surely people studying 110 with any seriousness have
already noticed this sort of thing, but I think it's really neat
that its "physics" supports a conservation law, namely phase
shift of "particles". In fact I suppose this is
probably extremely common in any CA with a periodic background...

Got up about 8:30, showered quickly, packed, checked out of
the hotel, and met up with a couple of other prospectives to
head to the CS building. Strange days, when I'm the guy-who-
knows-where-he's-going in a group, but I was, since I had
stared at the map a bit before leaving.

Got some standard-sounding spiel about the grad school, its
history, philosophy, etc. Thereafter more specific lectures
from Andrew Appel (PCC and TAL), Brian Kernighan ("little languages",
programming practice, AMPL), Vivek (Systemsy stuff),
David August (Compilers), Dave Walker (Linear logic),
Randy Wang (OS and Filesystems). Played with rule 110 during the
talks I didn't care so much about:

Phase shifts of stable patterns observed: +3, -3, -5.
Basic glider: (->) +6 (speed 2c/3) (<-) -6 (c/2)
Other gliders occurring "in nature"
(<-): 7 (4c/15), (may be more than one of these)
+1 (c/9), -1 (4c/15, rare?), +5 (4c/15), -4 (???)
(->): -3 (c/5)

Ate a little bit of the provided lunch. A bunch
of the PL people clustered at one table, incl. me, Evan,
Appel, and a girl from MIT (I think) named Rachel Greenstadt
who sort of monopolized Appel.
Left early to get a cab to the Amtrak station, thanks to Melissa
Lawson's last-minute arranging on my behalf.

Had an amusing conversation with the driver. I was trying to
explain what I wanted to do for grad school. He understood as
far as it was CS, and mentioned his son had transferred into
it. At one point he asked me if I ever "took any of that
foortran[sic]" Heh.

Got my ticket and grabbed a Roy Rogers burger at the
station. Had a bit of protocol confusion over the
delta between departure time and boarding time (answer:
much, much shorter than with an airplane) and concerning
when exactly when your, uh, safety certification is checked.
(answer: after (although not long after) the train is already
moving, which still surprises me) Apparently a business-class
ticket was purchased for me by the Hertz people, but I got into
one of the coach cars. Maybe I'll wander back to the other car
at some point to inspect its swankness.

Playing with 110 yet again on the train.
With the "empty string" (actually the perl frontend
fills in the standard background) on a 261-cell
universe with periodic boundary conditions, I get
a -6 (c/2) glider I haven't seen before together with
one instance of the usual +1 (c/9).
in 261-cells x............x offset 24 gives a rightbound +5 (c/5)!
Blank 262 gives a leftbound -4 (c/3). Maybe this is the -4 I spotted
before? in 262-cells x.......x gives a leftbound -4 (4c/15).

Hmm. Now that I get here, I don't see exactly what's so great
about business class. I guess there's a bit more leg room,
but it was already much better than an airplane when I
was in coach.

Need to remember to track down Ryan William's email address.

The taxi from Union Station to Arlington was even weirder.
The driver rambled for like two minutes solid at the start
about how I should have explained more carefully that
by "Arlington Hyatt" I meant the freaking -Hyatt- in
-Arlington-, as opposed to Crystal City. I swear,
I wouldn't think there'd be many different ways of saying
something so tautological, never mind the fact that
-I- never said anything about Crystal City, nor even
heard someone suggest that that was where I wanted to
go before the driver was already chiding me for doing
something wrong somehow.

I just saw a woman with a bag labelled "Alcohol
Addiction Recovery Month" go outside to smoke a cigarette.

Anyway, the interview went swimmingly. The interviewer was
a physicist and the questions were heavily biased towards
the sort of things I hardly consider to be mathematics,
i.e., calculus, diff eq, and matrix algebra. Oh well.

Got a cab back to Reagan, flew back to Pittsburgh.
I turned the wrong way upon leaving the gate and went
all the way to the end of the A-concourse before realizing
my mistake. I turned around and realized that my gate
was the very closest one to the center of the airport,
so the error could not have been more pessimal.

Took 28X back to Morewood, went from Morewood to Wean, and hung out
the lounges for a while. Thereafter, finally, sleep.