PhD Memoirs #3 : A Silver Lining

This post is being made with some emotional elements. Readers who wish to understand the below details fully are advised to go through #1, #2 entries.
 

I received a mail today. And it may well turn out to be one of the most blessed ones related to my PhD studies. Certainly I will cherish this moment. Okay I will not keep the reader guessing any longer.

After the discovery mentioned in #2 I decided to write a mail to the 2nd author of 007 (who, very fortunately for me, is still alive and I sincerely hope is in good health!!!), informing him of my discovery and requesting his own comments on it. Here are a few excerpts of that mail:   



Repected Sir,

Greetings!

... This mail is regarding some doubts that I have in Dr. Adnan Niazy's PhD work (1974) on the surface displacement field generated by a moving line-source double-couple model in a homogenous elastic half-space.

I have not been able to understand the particular method of solving the improper integrals adopted by him but I have been able to solve them using Garvin's approach ('Exact Transient Solution of the Buried Line Source Problem', 1956). However, upon obtaining their numerical values (using MATLAB) I am unable to get an exact match between my results and those (given in the form of different plots) of his thesis.

In the paper 'Discrete Wavenumber Representation of Seimic-Source Wave Fields' authored by you and Dr. Aki, it is mentioned that there are some errors in his results in the form of 'unreal high frequency content'. I have been able to get an almost exact match between your results (Figure 9 of the same paper) and mine (please see pages 1 and 2 of attachment). Since I have solved it analytically, assuming it to be correct, my guess is that the slight mismatch is due to absence of contribution from high-frequency terms in your results on account of the summation over finite number of discrete frequencies. Could you kindly let me know what you think of it?

.... Sir, could you also give some of your own comments on this issue or refer someone else who can help in this regard.
In particular, I am interested in knowing whether my plots (pages 3 and 4 of attachment) of the horizontal and vertical displacement field generated due to a buried line-source with step-function time history, are correct. Currently, they do not match with Dr. Niazy's (Figures 6,7; pages 63,64 of his PhD thesis; also shown in the attachment).

This is important for my work since my subsequent computations for a more complex fault geometry are based on those results.

Thanking you!

Sincerely,
Varun Kumar Singla

I sent it on this week's  Saturday (11/1/2014) and got his reply a couple of hours back today (13/1/2014). I wasn't expecting any of the sort below:

Dear Varun Kumar Singla,

         The figures you sent me for the comparison between your calculation and DWN are excellent. Congratulations for solving this problem analytically. I know it is a challenging problem.

      ...I think you should just compare your solution with the discrete wavenumber method which has been thoroughly tested, and unfortunately not rely too much on Dr. Niazy's solution which at the time could not be tested against another method.

With my best wishes for the continuation of your work,
Michel Bouchon

After reading it I experienced a feeling of achievement and some sensation of pride after a long time, I think or least do not recall having felt it to such an extent. After a while I was also filled with gratitude towards Dr. Bouchon for his timely and affectionate reply, which I conveyed him to him properly a moment ago. 

I felt like doing moonwalk like Michael Jackson but sadly didn't know how to. So I simply decided to listen to some of Hans Zimmer's 'Man of Steel' soundtracks and boy, they gave a very different feeling... quite different from those that I have been getting while listening to them almost everyday. I was on my bicycle, heading towards my lab, with a renewed sense of motivation when I was listening to those tracks, and suddenly I had this vision in my mind - I was standing, encircled by Batman, Superman, Henry Barthes (the lead actor in the movie Detachment), Jack Harper (Tom Cruise' character in the movie Oblivion), Arnold Schwarzenegger (with his Terminator looks and outfit), just like the scene in Harry Potter and the Goblet of Fire movie in which Harry gets to talk with his parents after getting his wand connected with Voldemort... and they all shook my hands in a congratulatory manner and had a smile (unique one for every character) on their face. 

I guess I am now almost back to feeling normal again. I still have much more rivers to cross I guess. The difference now being that I know what it feels like to know when a particularly dangerous one has been crossed and I hope it makes all the difference! 

My advice (yup, it does seem very presumptuous of me even to me, but I have a strong feeling about it) to all other struggling, wrestling, stumbling, dreary, restless, withered, etc. PhD students: 

1) Hold on. Don't give up easily without a fight...!

2) Listening to 'Man of Steel', 'The Dark Knight', 'The Dark Knight Rises', 'Oblivion' soundtracks may help.

3) By all means try to get help from outside faculty if facing problems. I hope some of them turn out as nice and kind as Dr. Michel Bouchon!

4) Inspire and help to propagate a culture of collaboration and cooperation in research, both as a student and later on as an academician/researcher. Your help, however small, may be of high value (or even priceless) to the recipient.

One final remark : sense of satisfaction achieved after getting appreciated for one's honest labor and hard-work is seldom possible to get by other means, I think.

Life as a tutor #3: Another beginning

I am writing this in a bit of a rush. 

My tutor job got over in November 2013 but due to some reasons I couldn’t write a blog post  then summarizing my experiences which I felt deserved to be done. But now there’s will be another opportunity for me to do that!

I applied for the tutorship of the same course this semester too. I had quite enjoyed that experience; it made my stay at IITK much more interesting and kept my mind active which, in turn, probably proved beneficial for my own research work by stopping my brain's neurons from slumbering. So, I decided to take up this task again and luckily got it this time as well!

As before, there are about 30 students in my class and the overall structure of conducting the tutorials is the same but with one change – the class timings! I took the first tutorial class on this week’s Wednesday during the early morning hours of 0800– 0850. And I will be following the same routine for all subsequent weeks as well. Also, all the main lectures being taken by the Course Instructor have been scheduled at the same time. If you are familiar with the weather in northern part of India during this time of the year I hope you will have some sympathy for me. At least I do, for the students! The rooms in which the tutorials are conducted are not warm and cozy. So, it will pose an additional challenge to my ability to teach and write on the black-board (I mentioned in one of the previous blogs how inconvenient I initially found writing on the blackboard but now, thankfully, I have adapted to it). Plus, the journey from my hostel to the class venue takes about 5 minutes on a bicycle and even in that small duration I get to know nature’s coldness quite well, and literally. 

But ultimately my major concern is how well I am able to teach and that whether the students feel relaxed enough in my class. So, I decided to relax some restrictions which the students generally feel obligated towards abiding by, such as:

( 1) Punctuality – I have given them full freedom to enter the class (and without asking my permission) at any time of their choosing. Of course, I will start the tutorials right on time.

( 2) Attentiveness – I have also given them the freedom to do whatever they like (e.g. sleeping) in my class as long as they DO NOT DISTURB OTHERS. In a joking sort of way I requested them to sit at the far end of the classroom if they had any such intentions.

And I considerably stressed upon the fact that they are free to ask questions. In fact, I consider this to be the biggest incentive for them to attend the tutorials. Of course I think that for many of them (though I hope I am wrong about it) the main incentive for attending the tutorials is maintaining a good attendance record for which 5% overall marks have been assigned by the Course Instructor. Either way I wish them to feel as comfortable as possible while sitting in the class. From my own experience I believe this makes learning much more easy and fun. 

I disseminated the above information in a sort of non-serious way to break the ice. It will be interesting to see their response and the number of vacant seats  in the class in the coming weeks.

Oh, and I shared the web-link of an online discussion forum with them which I remember sharing at a later stage the previous time when I was a tutor. Clearly, experience has its advantages.

I got to solve an interesting problem in the first class and tried my best to make it as interesting and understandable as possible. But I ended up not being able to finish it completely and so I would have to manage my time well in the next tutorial. Of course, some time was also spent on making the above announcements/introductions to the students so that I can hope that this won’t happen often.

Until next time!

PhD memoirs #2

At last I seem to have got some breakthrough. It turns out that t­he mathematical results which I have been using for reference and for comparing mine with (for a long time, after putting in much effort and concentration and with quite too many unpleasant experiences, I might add… won’t even go into the details) are themselves in error! This is unfortunate since not only have they been published in a reputed journal but also emanate from the PhD thesis work of a MIT graduate whose late supervisor was a person of high repute (http://www.usc.edu/dept/pubrel/trojan_family/autumn05/Aki.html).

Aah, do I hear you asking how I can be sure of it? Well, I found t­he ‘unreality’ of those results mentioned (rather, exposed) in another research paper, let’s call it 007. And I managed to get a match between mine and 007’s supposedly error-free results (the image below: cyan-colored curve is mine and the black one, 007’s).

 

Of course, it’s possible that both mine and 007’s results are incorrect and I am committing a sacrilege by questioning the PhD work’s integrity but it seems very unlikely at this point – (1) The odds of both 007 authors and myself getting the same incorrect results worked out using completely different techniques are very small (2) One of the authors of 007 is the aforementioned late PhD advisor.

Imagine that!

I wish I could say I feel very sorry for the MIT graduate student (i.e. when he was a graduate student, to be precise)…  I somewhat do... but I have myself struggled a lot at the cost of his errors, which I understand is probably not at all a result of his carelessness but simply a consequence of him being a human.

However, my first emotion was indeed that of elation. From here on, I can progress at a faster pace.. But I don't feel much cheerful. I know that it’s a relatively big achievement for me but I can’t help feeling a little anxious about the next course of action. I have spent way too much time on this and that has put a damper on any sense of celebration that I briefly had. 

I hope things look more interesting from here on, though some excitement does seem to be back.