All 2 entries tagged Wei-Chang
No other Warwick Blogs use the tag Wei-Chang on entries | View entries tagged Wei-Chang at Technorati | There are no images tagged Wei-Chang on this blog
February 11, 2012
I don't remember since when I have become a true believer of the Murphy's law, but I always see what it says happens around me, not always on me personally, again and again. The Murphy's law can be stated as, “Anything that can go wrong will go wrong.” It is not a real law in any scientific or philosophical sense, but it seems to be true when one looks into the history. Things usually go wrong in places one does not expect, and that includes your PhD study as well.
What can go wrong in your PhD study? A lot of things. You may have an accident during a holiday break and have to stay in a hospital for weeks, if not months. Your hard disc drive may fail, and you lose months of hard work just that. Your supervisor may accept a job in another university and plan to leave within six months. There can be loads of ways which potentially make our stressful PhD lives even more difficult. Although not all of them cause big problems to the PhD study, perhaps just temporary inconvenience and frustration, but someday someone could face a really devastating event, a zombie apocalypse, for example (though at the time we wouldn't care about the PhD anyway).
One of the reason behind the statement of the Murphy's law, I believe, is that we don't have full control of our lives. By saying “full control” it actually implies that we have the full awareness to things that we believe we are controlling. Taking my own experience as an example, my research involved some material that I had no way to produce, so I and my supervisor sought a collaborator to give us what we needed. In a review of my annual progress reports, the reviewer pointed out that to rely on a collaborator providing the necessary material might not be a viable plan. Then the reviewer suggested me to think of a backup plan in case I couldn't get the material.
Anything that can go wrong will go wrong. Remember that?
So this is what happened. Of course we understood the risk, but we thought we had had a positive response from the collaborator so that it would go smoothly. The problem was that this collaborator wasn't my supervisor's colleagues/academic friends; they didn't know each other. My supervisor just happened to meet him in a conference, and they talked a bit about the project I was working. Then they made a deal, sort of, for the material I would need in the experiment. A few months later, when I had gone trough all the tests and was ready to run the experiment, we notified him that it was about time to do the experiment, and we would need the sample pretty soon (giving him some time to “cook” the material). Then nothing happened besides several email exchanges, until I was running out of time and had to give up this material.
Even I thought I had a thorough plan for the PhD project, the Murphy's law still caught me when I wasn't prepared. In this case, apparently I knew the impact could be huge if the collaborator failed to deliver the material as he promised. But I chose to believe him in good faith, partly because I was convinced by my supervisor that the collaborator would keep the promise. Here was the blind spot. I believed I was controlling something that I actually wasn't. I overlooked the potential impact and prepared nothing for it. So this was how I almost screwed my PhD, well, one third of my PhD actually (see this previous blog entry). Remember the Murphy's law, always.
March 11, 2011
One funny thing about physicists, according to my years of observation, is that we often categorise ourselves into three divisions, theory, simulation and experiment, just as natural as breathing air. In each group they build up different expertise and perform their own practices. As a result of that, this categorisation actually makes much of sense. It basically represents the fundamental difference of methodologies followed by researchers in physics and, on a broader view, all other physical sciences.
Despite making such divisions, there is hardly any reason to believe this is THE way to conduct research. In the early days, a physicist would develop mathematics and theories and perform experiments altogether. As our knowledge having advanced, more and more physicists spend much time to ponder in a very narrow field of research over their lifetime. In the end, the specialisation turns every new physicist into one of the three categories in all but exclusive ways.
My question is, what would happen if you put the characteristics of all three methodologies on a single person? Imagine that a physicist can build a new theory, turn it into a simple mathematical model, test this model through numerical methods and prove all of them works nicely with the reality through experiments. Is it possible to train such a physicist in the modern world under our current infrastructure of higher education?
And an even more interesting question is, what would happen if this very person were under supervision of a professional physicist who had been trained only in context of one of the categories, say, a theoretic physicist? If the student's knowledge and skills were still far from mature in any direction, what could this supervisor do to achieve the goal, to make a new all-rounder physicist?
This is exactly what I and my supervisor have tried to achieve in the past two and half years. At the beginning, the reason behind this for me was just to do some brilliant research and to make my CV impressive. The academic jobs in physics, no matter a post-doctoral research associate or a lecturer position, are usually subject to in which category you are. Obviously, expanding my expertise to cover as many disciplines as possible would possibly add some more options when I look for a job.
It is not easy, though. Even I already had had a mixed background before starting my PhD, I usually fell into a biased thinking (mostly favouring theoretic explanations) against the problem in hands. For example, I would feel all right with some uncertainties about my experiment set up, as I knew the theory A predicted everything would go as I'd expected. But this is a wrong practice of scientific research which is explained by Richard Feynman's “cargo cult science”.
On the other hand, it is sometimes problematic to explain to my supervisor why his idea won't work. As a theoretic physicist, my supervisor expects something to happen as the theory A predicts. In such cases, I can't just tell him the negative result I saw but I also need to tell him where the complication arose and how that “broke the theory prediction” and so on.
Nevertheless, the whole experience is very positive, even though frustration takes a significant part of it. The greatest benefit of doing this is that it forces me to think problems from many different angles; sometimes the difference is big enough to make you think you're like speaking an exotic language to others. It is very intense to compress so many things, reading piles of mathematics textbook and journal papers, doing calculation, programming, long time working in laboratory, frequent travelling, etc., within a three-year PhD project, but I already feel the taste of what we have hoped to achieve. So I suppose that makes everything, good and bad, worth the effort I have made.