Travis Garrett's Blog

Really wonderfull space colony website. Lots of great paintings of the rounded rotating interiors of different designs.

Hmm, didn't really mean to post that, but needed to print out a progress report for my advisor and given the strange printer setup here that was the fastest way. Seems appropriate for a blog tho.

Travis Garrett: Spring 2004 Research Summary.

I am still working on my first large project, a 3-D computer simulation of a binary star inspiral using a co-rotating reference frame and a scalar gravity approximation (Watt&Misner gr-qc/9910032). I think the code is almost finished: the evolution for the individual spherical components works well (the hardest part probably being transparent boundary conditions - always some numerical reflection), and the synthesis code works well also. The primary problem is getting the proper evolution of the system since it depends on very delicate spatial derivatives (evolving using geodesics...). I am trying to get around this through several different approximation methods and I suspect I will be successful soon. When I am I will write my first paper on it, and go on to have my prelim. The PhD project will be on the same problem, only using tensor fields. I want to write at least 4 papers in my remaining 2 years here, at least a couple on the main research I describe here, but a couple of shorter ones also - talking to Martin Bojowald I have some ideas on evolving through singularities using a discrete mesh at the Planck scale (and as a relativitist I should have some grounding in quantum loop gravity, which I'm studying now), on a similar line I'm looking at Jack's work on quantum foam, might be good for a quick paper, and I also still have an interest in evolving large systems of particles - like globular clusters or colliding galaxies, and have some code written for those, might get a quick paper out on those also. I'm also 60% the way through a formal write-up of Statistical Metaphsyics, where I derive being an observer from all mathematical structures, although I'm not sure where I'd publish it (but Max Tegmark thinks I should publish it somewhere, perhaps a philosophy journal). In any case, I've got a lot to do.

I just noticed something interesting. For a certain amount of mass, if you compress it enough it will form a black hole with radius R=2MG/c^2, where M is the mass, G is the gravitational constant, and c is the speed of light. For the sun this radius is about 1.5 kilometers. For all the mass in the visible universe, the black hole radius is (drumroll), the current radius of the universe. More on this latter...

hee-hee

and a good nanotech page.

Cool Chinese Architecture

Interesting talk by Sean Carroll on eternal inflation and creating a low entropy situation at the big bang. Some inflationary models use a scalar field throughout space, then rolling down the potential hill of the field will deposit some of the potential energy into spacetime, causing it to expand. Then quantum fluctuations can take you back up the hill a bit, for even faster expansion, and even in a universe that has been spread out very flat by one inflationary period, one region of it can make another quantum jump and create another inflationary bubble. Any one finite region of space will thus get expanded into an infinite volume - going either direction in time no less! - and thus for every finite region entropy will increase indefinitely in both time directions. Thus our big bang with it's low entropy was not special - it was a perfectly typical and infinitely repeating situation.

Just wanted to add a little more to flying dreams: I had another dream when I was about 14 where I actually got to float in the air (even better!). Still it was only a foot or two in the air, and there was a trick to it - you had to focus on it very hard, and cross your eyes. I'm pretty sure the inspiration for that one was the magic eye stereogram books, where the computer arranged random dots would make 3-d images if you focused your vision either in front of or behind the page. Also interesting was that in the dream I was floating over my bed, and bright orange light was spilling in through the window. I then wake up and its morning and the bright sunshine is shining on me, so it was very disappointing - hey I was floating a second ago, and it was right here!

Now that I think about it, I have very fond memories of having my bed right up next to the window - the sun shining in in the morning is always nice (I always want the windows in my bedrooms to face southeast), but best were the cool cloudless summer nights with the full moon shining over the field of grass, almost bright enough to maKe out color. Maybe a few whispy cirrus clouds, and the night sky wasn't even particularly dark - more of a light grey. Just the cool air and the moon and a few stars over the grass.

Had an amusing dream last night - one could make small objects (glasses, pencils...) float in the air, if they were very finely balanced and placed in the correct position. In my dream I rationalized this by saying the centripetal acceleration from the rotation of the earth was holding them up - they were in geosynchronous orbit! This would be possible at the equator if the earth sped up so that we had about 1 hour days, which is much faster but not as if it would be flashing night/day like a strobelight. Likely dream motivation: I've been rereading the Hitchhiker's Guide before I go to bed, and am at the point where Arthur learns to fly by ignoring gravity...

More neural networks stuff. In Hebbian learning neurons will form random dendrite connections to each other, then the connections that are used frequently - they correctly model information - are reinforced, and the others for which the firing is asynchronous (the two neurons are not communicating) fade away. Iterate the process and you evolve a complex network.

Some cool pdfs on neuron cell structure from mni - the Montreal Neurological Institute.

Update - was being silly, scored a 15 on this test which is a little above average. Sure, its obvious for most of the questions which way they will sway the result, but I still answered truthfully.

New chess program Hydra is doing well against others, and uses FPGA - Field Programmable Gate Array. FPGAs are cool because AIs will need to be able to evolve their neural networks, as the brain does by varying dendrite connections, and FPGAs thus allow for hardware evolution in addition to software modification. Sure, all Turing equivalent, but FPGAs can be much faster in some circumstances.

Hmm. its 11 at night, and it turns out I am going to update my blog on its 3rd day of existence. Got a new CD player since the old one died after 7 and a half years (I fixed it for a week then it died again) - can get a 300 disk player for only 200 dollars, I guess because it is rapidily becoming archaic tech. Thus could put almost all my CDs in, and hearing some of the old ones made me remember softmore year of college and got me a little depressed, which is kind of rare. Actually read today on plastic about mind-reading - rather just picking up emotional cues from other people in their subtle facial expressions, and they not surprisingly said that autistic people do the worst at it, but more interestingly there is a sort of continuous distribution from them to the most sensitive. Made me wonder where I lie on the scale, and while of course I find my own emotional range and frequency to be perfectly fine and normal, I suspect I may be a little more to the autistic side. Maybe I'm being silly, but then I do prefer to spend most of my time inside my own head, thinking my thoughts - just 80% introverted I guess, although the 20% outgoing is quite...

Anyways, decided me and Katie should go out and have some fun, we saw Starsky and Hutch, which was ideal, and then messed around in barnes and noble for a while - read some of Smolin's new book and got inspired - I really need to work on both my physics and metaphysics very hard tomorrow. A dramatic storm blew through, the kind where little orange scroll bars appear at the bottom of the TV warning you about it, lightning and howling gusts of wind and everything, early on in March too.

Only a small selection of my experiences for the day, I guess the rest will fade into the depths of time, but then perhaps so will everything else when the stars all die trillions of years from now - in those small number of realities (but still very real) where complexity doesn't grow forever. Goodnight.

Programming DNA? Sounds pretty good for nanotech...

Been an interesting couple dozen hours - saw Lorenza'a day old baby yesterday, she had a full head of hair, was very quiet and observant, and had oddly capable seeming hands for a newborn, not that I've met many. There is the idea, perhaps with some validity, that the years seem to go by quicker when you get older beacuse each one is a smaller percentage of time compared to your whole life. Thus me, Mark, and Heather spent a couple months of subjective time with the baby by visiting with her for 15 minutes. And it seems reasonable, looking into its eyes, that their brains are forming connections at a million a second. Have too, have many trillions yet to go...

Had a couple beers that night with Mark, Heather, Constantine, and Gongpu at 'He's Not Here', out under the trees and night sky, and chatted about the big bang. For some reason, I'm getting nostalgic about it, which is ridulculous. Time keeps passing... As far as the universe is concerned, my life so far has been about a hundreth of a second.

Katie went to take her big test this morning, so I went for an adventure - I'd always heard that there was a bike trail through the woods to the east of campus, but I'd never been on it the 2 years I've lived here. Even though its been raining a good bit today, I figured that would just make it more fun, and it was. Its nice to get away from the constant stream of words and thoughts in ones mind, and they do leave dependably when you're riding a bike down a thin, twisting, steep, muddy, rocky, jutting-root-infested trail. A couple times thoughts would start to re-emerge more successfully when the going got easier, thankfully one that crystalized several times was: "continue to keep an eye on the trail you damn fool!"

Got a tiny little scratch on my leg and then some mud on that, and amusingly that was enough to get it a little infected, it stung and got red. Thankfully some of the thousand trillion cells that work together to form Travis thought to wash the leg with soap.

First post on my new blog - check out this cool email from my dad:

Hey Guys,

Mom and I saw some interesting wildlife behavior today at HH I thought
you would want to read about.

It was sunny and warm today (Sunday) if you weren't in the wind. We
walked and ran on the beach, and on the way back we walked along the bike
path that crosses to the other side of the road from the little bridge
by Morning Beach. You know the rectangular pond that is connected to
the canal that runs beside the path. Today, there were 12 pelicans in
that pond flying up a few feet and then splashing back down, over and
over. It turns out there was a huge school of mullet trapped in the pond
because the water in the canal was only a few inches deep. The pelicans
would rise up in groups of 3 to 6 and hit the school at the same time.
The simultaneous attacks must have confused the fish because the
pelicans kept on catching them. It was deliberately synchronized and
coordinated hunting, which I have not seen before with pelicans. At any given
time, about 3 of them were taking breaks on the bank to rest and digest.
Then they would switch off with others. The attacks were almost
continuous and each dive bombing took only a few seconds. We watched them for
half an hour and during that time they must have caught about 2 fish
per minute. They were doing the same thing an hour earlier when we passed
by on our way to the beach. There were hundreds, maybe 1000 fish in the
school. I think a lot of animals are smarter than we give them credit
for, birds anyway.

Dad