Posts

Showing posts from November, 2014

Three so far

One of the main reasons I'm writing everyday is so my English get better and so I can communicate my thoughts easily. I've been part of the amateur astronomy club at IIT Madras, Astro IITM, for the last 5 years and talking to the students who attend sessions has been one of the best learning experiences. They have helped me learn how to express one's opinion depending on the level of understanding of the other person. Going further, I'm interested in science popularization and a good starting point, as I see it, is contributing to popular science articles to magazines. So far, I have three small articles in magazines, two of which I've written by myself and one of which I contributed towards. The first two articles are on the research I took up in the summer of 2013 and in the winter of 2014 . The articles were published on astrobites , "a daily literature journal summarizing new astrophysical research posted to arxiv ". The third article was an articl

Our ability to speak

I've started reading the book 'Guns, Germs and Steel' by Jared Diamond . The author attempts to paint a picture as to why certain regions evolved more prosperously while certain didn't, why technological and cultural advancements occurred in one region while they didn't in another and in general, how our world came to be. It's a grand attempt by a single author in a single book but it's been an impressing journey so far. I'm still in the first chapter of the first book and my brain is already starting to go off on tangential questions. A brief mention of the fact that the ability to produce sounds to communicate set me off on thinking how such a complex system evolved from rudimentary animal sounds. Human learning is largely biased towards visual and hearing cues. While visual cues such as writing or a sign language can be complex and do impart a large amount of information, even before one learns the ability to write or read one develops the ability

you should start using git/github right now!

It's surprising to me that I haven't written about git on my blog so far! In fact, I have a small post on git on my github page  so I guess I confused this with that. Either way, for lack of a better alternative, I am going to ramble about the practical use of git again! In fact, I've taken it upon myself, now and a number of times in the past, to convince my friends and those around me to use open source tools. From unix to latex, from git to ipython, I've acted as a small-time open source evangelist, so to say... Coming to the point, git is a tool to keep track of the changes you make in a file. In fact, some of you might tell me that microsoft word has this functionality inbuilt i.e display the various corrections/additions made in a file at various points in the past. Well, yes, git isn't automatic and you will have to manually save (commit) the changes in a file. But git gives you the additional functionality that the files are available in a centralized ser

Gravitational lensing in astronomy and supernovae

Jumping right to the point, astronomers recently discovered a gravitationally lensed supernova . Simply put, a supernova is a star blowing up. They are some of the brightest objects in the sky. In fact, the most recent supernova in the milky way galaxy was in 1604 and the supernova was so bright that people could see it in the day for over three weeks. It is worth noting that this supernova is at a distance of 20,000 light years. Gravitational lensing on the other hand is the bending of light due to a massive object near it's path. Einstein's theory of general relativity was in fact experimentally proved when gravitational lensing was first observed as light from stars behind the sun bent around it to reach us. Galaxy and galaxy clusters are the major sources of gravitational lensing from an astronomical perspective. Coincidentally, if another galaxy or part of a galaxy (object galaxy) is behind this massive galaxy cluster closer to us (field galaxy), the object galaxy

H \alpha spectral line profiles

Starting from ground zero, the H \alpha (Balmer) emission line is observed at a wavelength of 656.3 NM. It is one of the most prominent emission lines observed in galaxies and it was one of the lines that I was looking for with Prof. Jeremy Mould  as part of a summer project. Over the course of our work, we found something interesting. Given the differences in inclination of galaxies (from our point of view i.e face-on or side-on point of view), differences in distribution of gas in the galaxy and the galaxy's rotation curve, the shape of the H \alpha line differs a lot! In the lab, it'd be a simple Gaussian function but in practice, it ends up being a sum of multiple Gaussian functions with different mean and variance. Interestingly, it might happen that the spectral line has two peaks instead of one. Think of a case where you are observing a galaxy side-on i.e one of the arms of the galaxy is moving towards you and the other, moving away. Using the same principle as Dopple

A hastily put post yesterday

I wanted to write about the code to estimate the value of pi yesterday. In fact I did write one but not to the best of my abilities. I was in a hurry and I couldn't push the code to github so I could link to the html file. Instead, I just hastily pasted the html code from the code to the blog post. Hence the distorted look. It looks way better in reality. I've talked about ipython before and how i one can generate static html files showing code, results and more HINT: $  ipython nbconvert --to html file.ipynb On a different note, I just got two new books for the month of December, what-if by Randall Munroe and Guns, Germs and Steel by Jared Diamond . I've already started reading what-if rather i'm through half of it. It's a very easy read and it's a very interesting one at that. I bought the book given the weird questions Randall Munroe tries to answer and the weirder answers he comes up with. I've been an avid reader of this webcomic xkcd  and I fo

Relevant code and results for yesterday's post on 'Estimating the value of \pi'

Image
[] In [58]: import numpy as np import matplotlib.pyplot as plt % matplotlib inline In [77]: a = 100 alist = np . arange ( 0 , a + 1 , 1 ) anotherlist = np . zeros ( iterations ) count = 0. iterations = 100 for i in xrange ( iterations ): tempx = np . random . choice ( alist ) tempy = np . random . choice ( alist ) if tempx ** 2 + tempy ** 2 < ( a ** 2 ) + 1 : count += 1. anotherlist [ i ] = count / ( i + 1 ) * 4. In [78]: print anotherlist [ - 1 ] plt . plot ( anotherlist ) 3.0 Out[78]: [<matplotlib.lines.Line2D at 0xa8b9f58c>] In [85]: a = 100 alist = np . arange ( 0 , a + 1 , 1 ) andanotherlist = [] for iterations in [ 10 + 1 , 10 ** 2 + 1 , 10 ** 3 + 1 , 10 ** 4 + 1 , 10 ** 5 + 1 , 10 ** 6 + 1 ]: #anotherlist = np.zeros(iterations) count = 0. #iterations = 100