I Scored a 93%!

Visualizing Context

Visualization is application. There is no denying from that. Until an approach of converting some data into graphical form is molded into something that can be applied to some real data to gain information about the data or convey the underlying property (or properties) in an easier way, the approach is useless. This has been recognized by many scientists working in the field of visualization.

In the context of atomistic visualization of atomistic data from a molecular dynamics simulation of liquids, simple atomistic visualization so far implemented in currently available applications fall far short of the actual requirement. Structure of liquid cannot be conveyed simply by rendering nice spheres and bonds using blazingly fast hardware and algorithm. The problem becomes even more complicated when the dataset contains thousands of atoms over thousands of time-steps. There are simply too many things that have to looked into simultaneously to gain an insight into the data and see what is going in the system. When the number of types of atoms in the system grows, the ambiguity in bonding criterion among different atom types creates another problem is representing them graphically.

So, far atomistic visualization has focused in representing the sphere is nice graphical form with very little support in visualizing the liquid system. Liquid systems are irregular and were not simulated at a large scale even till four-five years ago. With almost exponential rise in computational power and storage space, large scale simulation of liquid systems has become possible and now is being done regularly. Lack of support for visualizing various properties in currently existing applications, have resulted in many ad-hoc homegrown computer programs for visualizing the generated data.

My focus is to integrate many of the commonly used methods that are useful in characterizing liquid systems into one coherent system with easy integration of modules in future in mind. Another main objective of this work is to present a flexible user-application that can be used to visualize large number of liquid system without modification.

What is Time?

What is time?

When I was trying to model a time varying data, suddenly this thought came to me. What actually is time? I then googled. I found that I am not the only one confounded by this question. Since beginning of human intellect, philosophers were puzzled by time.

Is it an emergent property like color, smell, or sound? Single atom doesnot have color but aggregate of atoms can produce beautiful colors. Similarly, one atom doesnot have smell, but nasty odors can come out of aggregates of the atoms. So, is time like that? Recently, space has been shown to be an emergent property. David Gross in his 2004 Nobel lecture said that he couldn't imagine physics with time as an emergent property.

Then I rethought about the problem of time. How do we perceive time? What is the hallmark of time? Answer to both the questions is the same: change. Change makes of be aware of the passing of time. So, answer to the riddle of time lies in the origin of change. How does change happen?

In my last post, I ruminated about some fundamental rules that I thought were unbreakable. Now I say, those fundamental rules guide the mechanism of change. Those rules put the whole universe in a constant state of instability. Instability is the origin of change. So, those fundamental rules maintain the arrow of time. Stable objects on the other hand maintain status quo, hence do not experience time.

Breaking the Rules

Rules are made to be broken.

I have heard people say this, especially when disobeying the authority or law. It seems true too. It is easy to break some rules. But can we break all the rules? or are there any rule(s) that we cannot break? I have been thinking about this and I came to a conclusion that there are certain fundamental rules that cannot be broken.

Our world is build upon the rules. Right from the core of nucleus upto the whole universe, everything works according to the rule. Society runs according to rules. A person lives according to rules. Computer is built upon rules. Science is build upon rules. Mathematics is build upon rules. So, what are the rules that are fundamental?

In mathematics, any theorem is built upon some axioms that are thought be true. With that assumption in place, everything else is built upon those axioms. How to prove the axioms are valid? If something disproves the axioms, whole edifice of mathematics crumbles down. Similarly, our human society builds upon certain core principles. Those principles are like axioms in the mathematics. In case, somebody violates those principles, problem occurs. What are the axioms or core principles of the physical world? Everything thing must start on something. What sets the rules of interaction among different things?

I read some news about intelligent design and its conflict with the theory of evolution. At that time, I felt it was heretical to think intelligent design as any rational ideal. Intelligent design says that the universe was designed my some higher intelligent being. By this, the proponents of ID were trying to bring the god into the realm of science. It doesnot make sense. But it also doesnot make sense to think about the fundamental core principles of the world. In our world, when we simulate something we set the rules and let the system behave according to the rules. Who set the rules for the world?

Evolution of course can explain the 'evolution' but not the creation. How does something come into being? What are the rules for creation of life? How come the law of fundamental interaction is satisfied by everything? And my final question is, what happens if we are able to break the rule?

Self-awareness

Where does it come from? Will any of the human endeavors to make machine aware of themselves be fruitful? These questions came to my mind when I was preparing for my Computer Architecture exam. I haven't take any robotics or articificial course. But as far as I know, every computers currently in the world are modeled using John von Neumann stored program model. In this model, machine instructions are stored in the storage deviced connected to the computer. Those instructions are read one by one by the computer and 'executed'. Whatever the result, the computer doesn't have any interpretation for that. Everything is mechanical.

In last paragraph I put 'executed' within quotes. 'Executed' in this case means, the computer generates control signal to control the different functional units. Those control signals are immutable. They are preprogrammed either hardwired or microprogrammed. 'Executed' didn't mean the computer actually 'understood' the instructions and carried that out. Again, question comes what does 'understood' mean? Intelligent beings do not blindly carry out an instruction. First they 'reason' about that. If the instruction makes sense, then only they carry out the instruction.

Now, question comes, an illogical entity, like computer, that never reasons about the instruction given to it will ever be able to reason about its own existence. Unless, we change the fundamental model of computation, it is not possible. Let me repeat, I am unaware of the things going on in the field of robotics and artificial intelligence. It may be true that they already are using different mode of computation, apart from simply the programming languages. If I find out more about that I will certainly post them here.

And lastly, I put 'reason' within quotes in last paragraph. What do I mean by 'reason', here? How are we able to 'reason' whereas the machines are not? The 'reason' itself is human invention. It is possible that we are simply trying to assert our own superiority by inventing things we know machines aren't able to do. Or may be we are insecure about our own superiority? Are we actually superior over other beings? If we continue like this, another question comes to mind. How did we evolve to be able to 'reason'? I think if we try to reason beyond this, then we need to take help of meta-reason, theory to explain 'reason' itself, and I am not sure I will want to do that.

I have been wondering why there are no blog on GoogleBLOG about the Department of Justice's request for billions of search-data, some of them potentially containing personal information. There are multiple posts about new DELETE button gmail. But isn't the request more important than the button. I couldn't help wondering if the pet named GOOGLE is pet anymore. I have read in multiple instances than corporations doesn't have any scruples. GOOGLE might just have been an exception. I had hoped. But the information they gather about their users shows they are not any exception. They are not stand-out in the corporate crowd. In stead, they are just a part of the crowd, which just relishes to project its own image as an benevolent corporation.

Sorry Google people, you lost my trust.

This, actually instead of being technical in exact sense, is a reflection about the book I just finished reading and my own situation.

"Linked." That is the name of the book. Written by Albert-Laszlo Barabasi, a professor from University of Notre Dame, a physicist by profession, it is about the network theory. This theory, I found out after finishing reading the book, has been used in large number of scientific disciplines to explain phenomenon previously almost unexplainable or thought to be too much complicated. For example, metabolic network in biological cell, behavior of the Internet, relationship among people in a society, among many other things.

I finished that book yesterday evening. I wanted to write a small review right at that moment. But somehow I lapsed in that. Interestingly, this morning I read a news about a Cornell University Computer Science professor, Jon Kleinberg who recently received McArthur Fellowship from John D. and Catherine Foundation worth $500,000. That is a huge amount, especially when you receive it unannounced and suddenly. Over that, guess what?! He received that award for his work in Network Theory !!! That award is also dubbed as "Genius Award."

Here, I am trying to figure out way to find relationship among different types of atoms in a sub-thousand atom system so that I will be able to visualize is properly to project maximum information about their structural changes. People are researching about exploiting the inherent network relationship inherent among human-human, human-economy, human-society, and so on. Atoms also form networks. In this network, links can be their force-relationship, bonding-relationship, among many other things. But what characterizes a structure. When it comes to extracting properties about bulk-material from about hundreds to thounds of atom system, I am finding it really a daunting task. Over that, withouth proper training in physics, every little detail, is hindering my progress.

Anyway, good luck to Jon Kleinberg.