I play with algorithms to quantify the galaxy distribution with my supervisor, Andrew Hopkins, and other similarly enlightened characters. My honours thesis was a study of Extremely Red Galaxies (ERGs) in Phoenix, and here's the paper in which we also introduced an multi-scale structure identification algorithm. After much testing and development, we have applied this algorithm, Multi-Scale Probability Mapping, to the Seventh Data Release of the Sloan Digital Sky Survey. This work has appeared on arXiv, and it's pretty! And me? You mean me, personally? How nice of you to ask!**

JAN 2012 This may shock you, but I actually have some science to post here. Our algorithm, Multi-Scale Probability Mapping (formerly known as algorithm X) has been applied to SDSS DR7 to create a catalogue of 10443 galaxy groups and clusters and 53 filaments. Another way to think of this is that we have coarse-grained the galaxy distribution, providing a way to identify cosmic superstructures (filaments in the current application) as unions of groups and clusters. There are pros and cons to coarse-graining at this level rather than the galaxies; some information is lost but some of that information was noise, and noise can throw off algorithms looking for the non-trivial shapes of cosmic superstructure.

Details are in our paper, and pictures are here. As far as we know, no other algorithm has managed a false discovery rate of less than a half for filaments in real data, and we can do a pretty good job with voids this way too. I wrestled with the filament-finding problem for quite some time and am happy about how things have gone. Oh, and this is not the ``stroke of brilliance" I pompously foreshadowed last month, that's something else.

Er, and nor was finding these filaments the hideously stupid act I mentioned; that is a joke I'd love to tell but here's a secret I might actually keep!

SUN DEC 18 `Accepted 2011 December 15. Received 2011 December 15' plus sweet silence equals . . . something well that's just curious, I guess.

Star Control's Pkunk said that one's good deeds should be balanced by mischief now and again, lest one get so good as to turn completely evil. My version of this is that any hideously stupid act should be balanced by a stroke of brilliance. So here goes.

SAT NOV 5 Turning my attention to more practical matters, I recently learned that it's possible to submit all but one of the required copies of one's thesis electronically. To me, the only real advantage to a physical copy is its availability in the event of some kind of apocalypse, and the only texts people are going to want then are those that would aid in re-establishing civilisation. Knowledge of how to detect gargantuan but extremely distant cosmic superstructures is unlikely to help neo-caveman build his hut or jumpstart dead power plants, so I shouldn't be worried about printing it out . . . unless I can make this thing artistic enough to fuel decent fire-side conversation, or to be the intriguing subject of cave-paintings. Hmmm . . .

Current Supervisors (my lords, overseers, exalted effendis, . . .)
Andrew Hopkins
Dick Hunstead

Past Supervisors
Julia Bryant
Joe Khachan

Places
Sydney Institute for Astronomy, SIfA
My favourite genius, usyd's finest wavatologist and master of Kevlet Analysis, Kevin Aquino
The King of FAHB, Mat Guenette
My raja, Aditya Menon
Ben Jelliffe
Me

asmith@physics.usyd.edu.au Fear not the Arch Viles and Spectres of the Deepest Reaches, For the X is strong in this place. Last Updated - SAT JAN 7 2012

Star Control: Ultronomicon The galaxy teems with threatening monsters. May the Ultron be with you.