The ‘poster’ culture!

In tamilnadu, in every small town (and sometimes in the villages too), one would see posters wishing a young couple on their marriage, congratulating a politician on his achievements, welcoming some important person home from abroad, conveying condolences to a family on their bereavement, and so on and so forth. Sometimes the posters make very interesting reading, like the one I saw in my friend Shankara's village for the film star Ajith (who was visiting a nearby town for the mariamman festival, if I remember correct): It said, "As the presence of the pearls that makes the Ocean proud, it is the presence of our Big Brother Ajith which makes us proud".

This poster is one such (Courtesy Shankara). It welcomes our common friend Manivel Raja back home after his post doctoral stint in the US.

Here is Abi on some interesting posters that he saw in Srirangam, recently.

A paper and a story!

Long ago, among other things, Metallurgy Department at the Indian Institute of Science, Bangalore also had the distinction of being the only Department in the campus which ran a students' magazine. It was called "Metallograph". I used to contribute to it regularly.  Though most of what I wrote was of forgettable quality, here is a paper and a story that I wrote which might of interest to some of you.

Tips on doing a modelling and simulations research!

Recently, I completed my Ph D thesis. Looking back at what I did (or didn’t), I jotted down a few tips to myself. I am publishing them here for your perusal and/or comments. Like novels, research work also comes in genres — mine was a modelling and simulation work — I haven’t done any experiments. So, not surprisingly, the tips are about doing a modelling and simulations type of research. For experiments, you should look elsewhere.

1. Nothing is junk. Don’t throw anything away. Keep a copy of all data files, and plot files that you use to generate plots. They come very handy when you need to redo the plots. If, for plotting, you change the datafile structure, make sure that you keep a copy of the original ‘raw’ data — you just don’t know when you want to go back to the original data. Similarly, at the code development stage, do not throw away any data file that you may generate, even if you suspect/know that the data is wrong. Just label it to be known/suspected to be wrong. Sometimes, it might be right after all. Even when it is wrong, you can learn a surprisingly awful amount from your mistakes. Finally, the only time you want to throw things away is two or three years after you have documented and published your research — wait, even then I am not sure if you should throw them away.
2. Document the code. Reading the code once (after a fortnight or so of writing it) can help you correct lots of mistakes, and it is a mystery I have not yet solved: ‘If the code was so buggy, how the *@#! did it give right results?’ I have tried using cweb to integrate documentation and code. It didn’t work for me — not when the code is too big, anyway. What I have settled now is the following: one line comments with references to papers from which the formulation and/or equations are taken in the code itself, and a more complete doc file which is a tex file where the code pieces are included using \begin{verbatim}…\end{verbatim}.
3. Read, assimilate and throw. At the initial stages of my research, I xeroxed more than I read, and even now, the temptation to xerox is hard to overcome. In fact, for a while in the middle, I decided that I would xerox a paper, if and only if I have already read it with the result that in those days I neither read or xeroxed. Having said that, I realise that close reading of a few seminal papers is a must, and the earlier you do it in you research career, the better. At the end, all the papers go into the waste paper basket, and what we assimilate is all that we carry. Nothing wrong in referring to the paper for some detail — but the main ideas and lines of thought should be fixed in your mind, and, that is hard work which pays extremely well.
4. Keep a diary, journal, blog, whatever. It helps to jot down your ideas — most of the times, as you write it down, you realise how silly/stupid/useless the idea is. Somtimes, it takes a few days, and sometimes a bit of coding. But occasionally, the idea develops, takes shape in an entirely different fashion than in which it was envisoned, leading to nicer results, and those are the real gems. So, keep jotting down; the more useless ideas you are throwing away, the better would be that one rare idea that clicks.
5. Document, document, document. As soon as an idea clicks, write down a short note. Write down where the corresponding code, data files, and plots exist on your hard disk. Make the writing as detailed as possible. It comes very handy in pursuing further ideas along the same lines as well as in writing the stuff up for a paper.
6. Share — there is no better way of spotting mistakes. It is too difficult to make others go through your code, looking for mistakes. One reason they might do it is when they realise that your code can be used by them for their research. When they take a look at the code, they will come up with lots of queries, clarifications and bug reports. Such sharing of codes would also help increase the credibility of modelling and simulation work; and, more people will have faith in your results.
7. Prepare several presentations and practise. I have at least four different presentations made based on my thesis work — one for one hour talks, another for half-an-hour talks, another for twenty minute presentations, and the last one for five minute presentations. It is a really huge effort preparing a nice talk, and there is no point in preparing them everytime anew. Make a directory — keep all the relevant figures and plots scanned. Keep your presentations also ready. After that, every talk needs just a few hours of preparation and a few more hours of practise instead of days.

So much for now — if I think of some more things, I will do a follow-up post. Till then, have fun!

On how not to write travelogues!

Here is a very enjoyable article by philobiblion on a travel book that she didn't enjoy.

Basic Instinct 2: a review!

We watched Basic Instinct 2 the other day. The movie is disappointing and not worth the watch. Instead, check out Basic Instinct from your nearest video store.

A few hours of bliss!

That is what we experiencd yesterday in Neyveli Santanagopalan's concert at Seshadripuram Rama Seva Samithi. The concert started with Viribhoni varnam.  This was followed by Mayamalawagowla (Maayaatheetha swaroopini), Chaaya Tarangini (Krupa Judachu), Poorvi Kalyani (Gnaanamosagaraada), Saama (Annapoorne), Aanandabhairavi (Mariveregathikkaveramma), and Kalyani (Aethaavunara). After a brief interruption (about which I write below), he sang some nice thukkadas — karpagame, Thungaatheera, saarasamukhi… On the whole, it was a wonderful concert and very pleasant and relaxing.

The only not-so-nice part of the concert was the interruption to garland the artists, at which time, one of the organisers read out an obviously prepared speech, which was jarring in my opinion and spoilt the nice ambience created by the artist. I think it is time that we stopped these 'mukhastuthi' (as it is called) and get on with the music — the applause at the end of the songs and the exclamations from the audience during the singing is sufficient. And, probably, the artists should be garlanded at the beginning and be left to themselves and their music for the next few hours.  

In praise of Amitav Ghosh!

Here is a review of In an Antique Land of Amitav Ghosh. Like the reviewer, I too have not read any of Ghosh's fiction; but, his non-fiction writing is marvellous. In addition to Antique Land and Imam and the Indian that the review mentions, his Dancing in Cambodia is a truly-not-to-be-missed piece.

Juicy, thy name is gossip!

I guess it is Miss Marple who once made the observation that "there is so much of pleasure in malicious gossip". Here is Abi on gossip in academia and the role of gossip in social bonding. Go forth and spend an hour or two in the coffee house gossiping — you are just being social with people and serving the academic community

Pedas and music!

Not necessarily in that order Here is Ram Guha on the greatest Indians who strive and achieve perfection.

Teaching symbolic thought to apes!

Go here and read this post and this article to which it links; a certainly-not-to-be-missed piece.