Sunday, January 10, 2010

Right to Information

I read Times of India of January 8,2010 and there was a news which I had been waiting for at least 3 years. The title of the news was - 'After judges, babus' assets come under RTI'. I think that it is a landmark judgment which has widen the ambit of Right to Information (RTI). Let me quote the crux of the judgement - After politicians and judges of Supreme Court, now the assets of babus have been prised open to public scrutiny. In a landmark order, the Central Information Commission has said that disclosure of information such as assets of a public servant, routinely collected by the public authority, should be made available to the public under the Right to Information Act.

I believe that this judgment is going to reduce the widespread corruption in India. I think that Laws such as RTI in by themselves can't empower the public unless there are some clear consequences of the law shown to the public. With the way laws are written, there is always multiple interpretation possible if one can at all comprehend them. I am not saying that laws should be made simpler. What I am saying is that not many people understand what a law can or can not do. To make it clear what a law can do such judgments are needed. Once such judgments are made, they say clearly what people can achieve with the help of a law. It is the application, not the underlying theory, that gives us actual feel of things. It is engineering, not the underlying science, that is changing the face of the earth. It is these judgments, not the underlying law, that mark the revolution.

I have seen things changing in India due to RTI. My own uncle who is an Assistant development officer has been affected by RTI. In a week he has to go at least couple of times to the Information commision's court. He has to go to court when his department fails to provide some information to the petitioner. Although the punishments for not being able to provide the information are fairly mild (10-25k rupees), the process of appearing before a court and possibility of getting rebuked by judge are good reasons to provide information and perform duties scrupulously.

RTI is such a powerful tool that, when used properly, can lead us to the direction of rapid development. I remember Rajiv Gandhi's statement, ' If center sends 100 Rs for schemes only 15 rupees is actually used for the purposes of scheme. Our corrupt bureaucrats and politicians eat the remaining 85." I think that with proper use of RTI we can make sure that 85 rupees is actually used for the purposes of the scheme. So people come out of the closet and file applications under RTI to correct these corrupt people. Now that you have a land mark decision, you should not hesitate to peek into the private assets of government employees.

Saturday, January 9, 2010

It is all about not loosing your head

I played tennis after a week today and against an opponent of comparable skills. I realized that there was a moment in the set where I was leading the set with 4-2 but I ended up loosing it 4-6. After the customary hand shake we started discussing why I lost. The opponent said, "It is all about not loosing your head." He suggested that I was not in well control of my shots towards the end of the set i.e.I committed too many 'unforced errors'. I agreed to him on this and I continued the thought process after showing him my humbleness.

I thought about myself 3 years back when I just finished my undergraduate. Loosing in anything was so hard and it gave so much sourness and distaste despite any shown humbleness for defeat. Why is that I am not that sour or unhappy today after loosing. I guess in 3 years I have learnt loosing with grace. With these well timed defeats, I learn more than I ever can with a success. It just gives such a humanly chill. In-fact, I enjoyed congratulating my opponent.

Thursday, December 31, 2009

Adieu 2009!

Another year, with mixed happenings and experiences, spun by. I woke up this morning and realize that the last three digits of the year I wrote for a year will not repeat itself for another millennium. I realize that Dec 31st, 2012 is just 3 years away. I realize that tonight there will be a time point which I can count in 2009 as well as 2010 - 12:00am. Then, I thought lets have some written testimony about the year 2009, which I may not be able to remember as vividly later as I remember right now. Let me write some things from 2009 that I think are worth remembering in chronological order:
1. January-09
---A miserable winter at Madison (Wisconsin) with average temperature of 5 degrees Fahrenheight.
---I repent my decision of not going home in winter vacation of 2008
--- I decide that I will go home in March-09
2. February-09
-- My prelim report due
-- A month of researching and writing
-- Found out how incompetent a writer I am
3. March-09
-- My prelim defense where I was supposed to present my work of 1 and a half year
-- My belief, that I am a good presenter and defender, gets shattered
-- 2 eventful and colorful weeks in India
-- Good end of the month - passed my prelim
4. April-09
-- Month of reconciliation and introspection
-- Realized the importance of Marketing even in research
5. May-09
-- The learnt Marketing skills start to blossom
-- Gave an amazing presentation at the end of which found myself at self disbelief - Was that I talking?
6-8. June-09 - August-09
-- Learnt a new language - German
-- Now, I can answer the question "Who are you" with "Ich bin Rishi"
-- Got my India trip finalized during December-09 and January-10 with a very good airfare deal
-- One of my close friends' wedding got scheduled in November but I thought that I would not be able to go because of my preplanned India trip in December-09 and January-10
9. Sept-09 - 'The month of happenings'
-- My sister's wedding got fixed for November-09
-- Had to reschedule my India trip during November
-- The rescheduled itinerary made attending my friend's wedding possible
-- Everyone in the family was exhilarated for the upcoming wedding event
-- I can not remember me being happier than this ever before in my life
10. Oct-09
-- My total call time to home increased by a factor of 10
11. Nov-09 - 'The month of experiences'
-- Two weeks of India trip
-- Was in the organization side of a wedding instead of being an observer first time in my life
-- Was in the lead roles in some of the wedding ceremonies like Tilak first time in my life
-- A lot of fun in reunion of old college friends after 3 years at my friend's wedding
12. Dec-09
-- Back to the reality
-- Writing this post on December 31st 2009

By definition, any event that is not in the present year, is the part of history. Now, I realize that all the events that have happened in 2009 will soon become the history. But I am happy that I will have this blog to look back at the history if I ever feel nostalgic.

~Rishi

Monday, December 28, 2009

Is there anything random in this world?

A, B and C are discussing about randomness. Person A asks person B," What is the probability of obtaining head when you flip an unbiased coin?" Person B says, "If the flip results in head the probability of head is 1, else it is 0." A and C start laughing on this. But suddenly C changes side and says that he supports B. In-fact, C says that the coin flip is not at all random. C says that we can predict the result of a flip if we know enough information - height of the coin from ground, amount of force applied to the coin, the point where the force is applied and so on. Now A got C's point. Then A asks, "Why do we say that there is 1/2 chance of obtaining head?" To this C replies, "There is no way to find out that the probability of flip is indeed 1/2." A crazy scientist once decided to prove that the probability of head from a flip is 1/2 and he kept on flipping the coin all his life but was never able to prove this.

Laplace never believed in randomness. Here is something directly taken from wikipedia:
Laplace strongly believed in causal determinism, which is expressed in the following quotation from the 'Introduction to the Essai':
"We may regard the present state of the universe as the effect of its past and the cause of its future. An intellect (Laplace's-Daemon) which at a certain moment would know all forces that set nature in motion, and all positions of all items of which nature is composed, if this intellect were also vast enough to submit these data to analysis, it would embrace in a single formula the movements of the greatest bodies of the universe and those of the tiniest atom; for such an intellect nothing would be uncertain and the future just like the past would be present before its eyes."

The quotation challenges everyone who talks in terms of probability. Let me give some examples where probability may or may not be a useful way to think about scenarios:
1. "There is a 70% chance that India will win"
Now realize that Laplace's Daemon would know every thing required to predict India's fortune. He will know what are the mental, physical, psychological and spiritual states of all player, umpires, referees and audience. He will know ground conditions, weather forecast and so on. He can indeed say that at the end of the day whether India is going to win or loose. However, predicting India's fortune is very hard for even Laplace daemon because the amount of computation power he may require could be huge.
2. "There is a 90% chance that the bus will not arrive at the stop on time"
Now, amount of information required to predict whether the bus is going to come late is not huge. You may just have to call the bus station people, see weather forecast or even now a days a real time monitoring of status of buses is possible. So, if some one tries to use the probabilistic statement for making a decision, as to when the bus is going to arrive on the bus stop, he is not making a wise choice.
3. "There is a 70% chance that I will be late tomorrow for office"
Now this statement is totally absurd. People are so occupied with probability that they consider themselves to be unpredictable.

The three examples that I talked about have a salient gradation among them. In example 1 the problem of prediction was hard. In example 2 it was not so hard. In example 3 it was very simple.

Conclusions: At time saying things in probabilistic terms may sound smart at the outset, but can be totally stupid or absurd in the core. I personally think that prediction by probability should be avoided if possible for the simple reason that there is not a very foolproof way to estimate probabilities. Even if there were a foolproof way, probabilities can only make predictions which are probabilistic. It may so happen today that the bus arrives at bus stop early even though it was 90% probable that the bus is going to come late.

Saturday, December 26, 2009

Human communication

There is no doubt that among all species we know about, we are the ones possessing the best communication method - our languages. A friend of mine ones said that he always thought how on earth the first word describing an object has ever come out. Let us not worry about how it came out but let us say that the first word was 'ox'. An intelligent person sees a particular instance of a species and she decides that it should be called ox. Now realize everyone else in the world has no name for ox other than the person who I just described. Let us understand the problem this person faced:
1. I can call it ox, but can anyone else even pronounce this.
2. Who should I tell about it first. If she tells it to a few morons to start with, they would not care about her finding and may not even appreciate the need for a platform for communication. And every times she fails, her enthusiasm about her finding deteriorates (after all we are human beings). So she has to choose first few persons judiciously.
3. Let us say she has chosen first few persons who understand the importance of her finding. Now, if she has convinced these few persons that a particular type of species should be called ox, how every one else is going to realize that an ox is an ox.
4. This is where she realizes that she needs good artists - artists who can draw an ox in a wholesome way. Let us say, she, the discoverer of name of ox, finds a few good artists. These artists can draw a species which has a particular look and they know that it is an ox.
5. Now, a few people can call an ox an ox and a few other can draw ox in a proper way. But realize how inefficient it is to draw an ox and then tell people that you should call it ox. This is where she might have realized that we need a written form of ox that maps the picture of an ox to a few word with few letters. Let us assume that the first language was English (You can assume whatever you want, but I don't have many choices :D). Now, to write an ox she needs some building blocks.
6. The importance of building blocks (letters) can not be neglected. Let us realize that this person knows that their are other things other than ox which are to be mapped from their physical reality to some written representation. So she decides that we need an alphabet system. An alphabet system which has enough variety in it so that it can capture a lot of things but this variety should not be overwhelming (Our brains have memory limits :) ) . Now this person with the help of other persons has developed this alphabet system and defined written representation of all species.
This part, mapping physical objects to written form, was easy. I can only imagine how complex it is to define certain other nouns like passion, desire, hope, love etc. Mapping of these nouns from what is their in mind to the written words must be really complex. But somehow we all seem to understand all these words. I still feel that their are a few problems with the way we communicate, let me mention a few:
1. The problem of scale: Let me start with an example. Some one says,"My son has obtained really good grades". Now there is some grade that someone's son has obtained and according to that some one these grades are good. Now if this statement is ever told, I am almost certain that listener does not know what are the grades of the speaker's son. The reason could be that according to speaker anything above a B is a good grade but the listener may believe that nothing other than A's is a good grade. So at the end of the statement, the speaker and listener are not at the same information level.
2. The categorization problem: Two different persons can put the same object/event/thing into two different categories. An example would be blue whale. If a person with not much biology background is ever asked what type of a species a blue whale is, he is going to say that it is a fish. But biologists have categorized it into mammals. The categorization problem manifests itself, when two people have different definition of the same category. This problem is abundant in our communications. For example - some people say that India is democratic whereas certain other say that it is socialist democratic.

Conclusions: Humans are complicated. Different words have different meanings to different people. We can only hope that what we are saying is actually what the person in-front of us is getting. To me the only solution to this problem is adoption of a predictor-corrector method. Some one says something to you. You predict in your mind what he or she might mean. Now you ask back some questions and based on the replies, you correct the meaning in your head. The only problem with this method is that there is no guarantee of convergence i.e. you actually being able to decipher what someone meant.

Thursday, December 24, 2009

Why should we not rely on email?

Consider a situation. It is a Friday after noon. You (A) and your friend (B) decide to have a dinner together in the night. Both of you decide that you should dine at Little Sisters (LS). You two decide a meeting time 8:00 pm at LS. A and B like to talk to each other but the problem is that both of them don't like being the one initiating a phone call. So A would prefer writing an email rather than calling if s/he wants to communicate a small piece of information to B. At 6:30 pm, A gets to know that LS is closed that night. Now A writes an email to B that they should dine at Punjabi Dhaba (PD) instead of LS. A reaches PD at 8:00pm and waits for 10 minute but finds that B doesn't show up to PD by 8:10pm. Now A starts thinking. --- Did B ever receive my email. If B did not receive the email then he would go to LS. Should I go to LS and from LS take B with me to PD. What if B actually did receive the email but is late in reaching LS due to some other reasons. In that case I should wait for B at PD----. A also thinks about what B might be thinking, if B has gotten the email. --- B knows that their dining place is changed to PD. But B doesn't know that A knows that B knows that their dining place has been changed ---- .

This is the famous problem of information not being a common knowledge. A and B may both know a fact(F). But does A know that B knows F. If A doesn't know whether B knows F then they don't share the same common knowledge.

This is what happened in the dining plan of A and B. B got the email from A. So both A and B know that their dining place has been changed to PD. But A doesn't know whether B knows that their dining place has been changed to PD. Similarly B doesn't know whether A knows that B knows that their dining place has been changed to PD.

Conclusions: Don't write email to plan dinners :D. Email where the sender is not sure about receiver actually receiving the email should not be written at all. Use your phones people - a bilateral communication device.