Entropy is big lately. A new paper on arxiv (The Entropic Lanscape, Bousso and Harnik) uses entropic principles to derive predictions about such things as the cosmological constant and the nature of entropic radiation. These predictions match well with observations in our corner of the universe and give a framework that applies to any part of the universe, even in other vacua. This paper relates to examinations of the vacuum lanscape that usually end up relying on anthropic arguments. In other words things are the way they are because we wouldn’t be here to observe them if they were different. Bousso and Harnik replace arguments like that with ones that favor the maximiztion of entropy.
Also in a comment to a previous post here Nisheeth points to his arxiv article (The relativity of theory, Nisheeth) where he describes a framework for deriving physical laws from information-theoretic first principles. He too relies on the maximiztion of entropy.
Lately I’ve seen a several papers describing how one fundamental way of measuring the universe or another is actually emergent from a different way of looking at things. There are those that claim time does not exist in any fundamental way, we are just looking at things wrong. So what is fundamental?
Entropy.
Understand entropy and gravity emerges. Understand it in a moving frame and General Relativity emerges. The idea is published on arxiv (On the Origin of Gravity and the Laws of Newton, Erik Verlinde) and even made it to Slashdot. Woit surveys other entropy related ideas.
As a computer scientist there is something appealing in the notion that information is fundamental and that all the laws of nature are derived from the way information works.
Update: There is a long post from Verlinde on Motl’s blog here. It illuminates some of his ideas and explains their history. If you can stomach Motl’s bullying read the comments too, there are many.
This holiday break I started playing around with Mahout, getting it started and running some of the sample data through it. It’s a new part of the Lucene/Hadoop project in Apache which contains a math lib and code generator which build mapreduce jobs that run in Hadoop and use the HDFS to store data. It includes a nice vector and matrix library that provides a flexible set of operations and collection types implemented on the parallel mapreduce architecture of Hadoop. Included are several higher level frameworks of general purpose usefulness like:
a clustering engine using k-means clustering
a Breiman decision forest engine
a document classification engine using Bayes and Naive Bayes
a collaborative filtering engine
cosine similarity via vector dot products
In a previous job we implemented term vector based clustering and similarity in an innovative application for browsing an information space, which was calculated from a number of web pages. We built a prototype on Hadoop and Hbase and were in the process of moving the calculations to mapreduce when the funding ran out. It sure would have been easier with Mahout.
Motl has a nice post about the place of thought experiments in physics. How do you run an experiment when you have to have a train that moves at the speed of light and is smaller than the Planck scale—run a thought experiment.
Cosmic Inflation attempts to explain why the universe is so uniform by positing that the universe may have expanded faster than the speed of light, at least for a time. It is a very successful theory which allows us to think of the forces as having not changed with time. But what if Gravity itself has changed with time? What if the early universe had little or no gravity and later, through a sort of phase change (like water turning to ice) gravity was switched on? Could that be an alternative way to explain the state of the universe today and if so how might it have happened? Those are the questions that Brian Greene, Kurt Hinterbichler, Simon Judes, and Maulik K. Parikh try to address in a new arXiv paper.
The basic premise is that things are uniform now in the universe and if the universe had expanded at a constant rate with gravity acting all the time the laws would predict a more lumpy universe. Cosmic inflation questions whether the universe expanded at a constant rate. If the early universe expanded faster that the speed of light then the clumping effect of gravity would have been frozen out for a time. Then when expansion slowed down the effect would become more pronounced. As an alternative the new paper questions whether gravity was uniformly powerful the whole time. If a gas is unaffected by gravity the gas will distribute uniformly in space but when gravity is added the gas atoms will eventually form into clouds, condense into stars, and collapse into black holes. The new paper suggests that gravity itself might have been weaker for a time, leading to less clumping.
We tend to imagine our surroundings on geological or cosmological time scales as being governed by gradual change. If things do change the changes must certainly be brought about by constant laws. We think of the speed of light and the strength of the forces as being constant. But as we are able to see at greater time scales we have found over and over that things we thought were constant do change—stars die, galaxies form, hot rocks flows like water, and the universe expands at an accelerating pace. There are a growing number of people asking whether the laws of nature also change. Until we ask this we aren’t done looking at the possibilities so I was excited to read the paper.
Sean Carroll of Cosmic Variance has another great post about an arXiv paper that raises questions about GR. The gist is that the curvature of space and time should be equal according to GR (though I’m not so sure about that one) and looking back in time using gravitational lensing shows them to be different. Sean explains it best here.
Swarm is a framework in Scala being developed by Ian Clarke (the guy who did Freenet). It make use of Scala’s support for “portable continuations” in a novel way to spread code among distributed machines. Using Swarm you write lines of code, flagging where you want them to run, and all of the code and data is moved automatically to be executed in one virtual sequence. Nice but that is only the beginning — Ian describes using self analysis of execution patterns to optimize where the code runs even without programmer intervention. Now that would be a trick. He needs help so check it out.
Cosmic Inflation states that for a small amount of time near the big bang the universe grew faster than the speed of light. This inflationary period explains why the universe is so uniform. But what is so special about this universe? If the big bang came from a nucleation site (sort of like how the foam in a bottle of coke nucleates on the Mento) then there may be more universes. So if there are more universes out there what would happen if they collided? If we saw this happen what would it look like? This train of thought describes one type of multiple universe theory and in an area that has few theoretically observable effects the collision scenarios are worth thinking about. Besides the arXiv article by Anthony Aguirre, Matthew C. Johnson has some nice gratuitous graphics and some show-off math.
Spoiler alert: there may be cases where a collision does not anihilate us all, which means we might see the effects of the smack-down and it might be visible in the CMB as seen through the Planck observatory. In fact if you look closely in the picture to the right, well no you are probably going to have to wait for more to fill in.
If the early universe expanded faster that the speed of light then the clumping effect of gravity would have been frozen out for a time. Then when expansion slowed down the effect would become more pronounced. As an alternative the new paper questions whether gravity was uniformly powerful the whole time. If a gas is unaffected by gravity the gas will distribute uniformly in space but when gravity is added the gas atoms will eventually form into clouds, condense into stars, and collapse into black holes. The new paper suggests that gravity itself might have been weaker for a time, leading to less clumping.
Cosmic Inflation states that for a small amount of time near the big bang the universe grew faster than the speed of light. This inflationary period explains why the universe is so uniform. But what is so special about this universe? If the big bang came from a nucleation site (sort of like how the foam in a bottle of coke nucleates on the Mento) then there may be more universes. So if there are more universes out there what would happen if they collided? If we saw this happen what would it look like? This train of thought describes one type of multiple universe theory and in an area that has few theoretically observable effects the collision scenarios are worth thinking about. Besides the 
Spoiler alert: there may be cases where a collision does not anihilate us all, which means we might see the effects of the smack-down and it might be visible in the CMB as seen through the Planck observatory. In fact if you look closely in the picture to the right, well no you are probably going to have to wait for more to fill in.