SpaceTiger said:
The colloquium will be held Tuesday at 4:30 PM in the Peyton Hall auditorium. My thesis is titled, "Crawling the Cosmic Network: An Exploration of Filamentary Structure". I talk about ways of finding and quantifying large-scale structures in the distribution of galaxies.
I had the pleasure of attending this colloquium and I will give my account of it, not so much to inform you as to enrage the speaker to the extent that he is forced to correct the errors in it.
The room was filled with mostly graybeards, but there was this old German woman with a little too much testosterone. I will have more to say about her shortly. There were also a large number of young female attendees, We can only guess at what attracted them to this talk. Dr. SpaceTiger enthusiastically tore into his subject matter, the stars that shine and twinkle at night like so much candlelight in the eyes of these pretty young scientists. But not in so many words. It went more like this.
When you look at a 3D graph of the galaxies in the Sloan Survey, you see things like clusters, filaments, and walls. Indeed, he had an actual filament at hand to show us and which did double duty as a pointing device. These structures may be compared to the bunnies and horsies that we see in the clouds. But with this important difference. Dr. SpaceTiger was able to write some pattern matching software that actually found these structures in the data. Here's how he did it. First, he graphed the matter density, tossing in some dark matter here and there. Then he took the Hessian of the density (I'm pretty fuzzy on this step), and then he diagonalized the Hessian. This results in 3 numbers, the eigenvalues of the Hessian. Then he mapped the eigenvalues in such a way that the structures could be identified. To me, the most revealing image he displayed was one of the galaxies and with red lines where the software had found the filaments. Nice matchup. He showed how at the given level of contrast, it was easier to find filaments than walls. He showed how at redshift z = 3, there were more walls, and at z = 0, more clusters. He indicated that this means as time goes on, walls deteriorate into filaments, filaments into clusters, and given time, our local group will become a wooden nickel.
Then came the questions. Mostly these graybeards wanted to know why Dr. SpaceTiger didn't adjust this variable or that one. I really thought they had him there, but he always came back with a sharp retort that put them in their place. However, the evening was nearly ruined when the German hag piped up. She said that she got her PhD 90 years ago and she did some bogus studies that showed that these walls were very real indeed. I suppose she thought she was being oh so clever by pointing out that it is called "The Sloan Great Wall", not "The Sloan Great Filament". Actually I thought so too. I would have been in tears at this point, but Dr. SpaceTIger quoth as follows. "Up yours lady, I never said there weren't any walls, just that the filaments were more prominent at this level of contrast". She was speechless at this point. Dr. SpaceTiger ended his talk to great applause.
One guy asked if the filaments formed a kind of mesh like a tennis racket. This raised a question in my own mind, but having somewhat below the average knowledge of this matter, I decided to keep my trap shut. However, I will ask the question here. Is it like a tennis racket with a lot of snot on it?. I mean like a wall but with dense strands running through it?
. 