Friday, December 11, 2009

Why Sarah Palin Is Right*

(Originally A Call For Inaction)


Related Post

For long I get a uncomfortable feeling whenever the topic of discussion turns to global warming (also known as climate change, or Al Gore's lamentation) .  Problem is I don't know about science of climate change, and while arguing from point of ignorance is fun, such a debate doesn't serve any constructive purpose, unfortunately of course for many people entertainment of pissing contests is the point.

Now my ignorance of climate science doesn't surprise me, the things I am ignorant of vary from string theory to cooking vermicelli. However unlike string theory or cooking vermicelli, climate science affects public policy which is why some sort of understanding is required. Now I admit that when I had become skeptic of climate change, it was partly a fatigue to incessant, and one suspects grossly exaggerated, alarm about global warming.

But if global warming is crucial for deciding public policy, some sort of understanding is essential, and yet climate science is too complicated a subject for someone like me to understand in his spare time. This being the case, I was stuck in undecided category.

Then sometime back it struck me I was approaching the problem from wrong angle. The correct approach, since the question involved public policy, was to do a cost-benefit analysis. Again since I don't understand climate science, I can not estimate the cost of global warming. As it turns out, it is much simpler to estimate the cost of global warming mitigation, which I will try to explain in as complicated terms as possible,

For calculation, I am assuming that  Al Gore is right and carbon dioxide indeed causes the global warming**, and global warming is a clear and present danger*** (and by clear and present I mean around fifty years on geological scale). Further it is reasonable to assume that warming on global scale must be lagging indicator of  percentage of atmospheric carbon dioxide.

This implies that effect of a reduction in carbon dioxide will be visible in terms of reduced temperature only after certain period of time, let's say thirty years (considering the geological time scale it errs on the side of caution)

 Now if both the previous assumption statement are correct, one might conclude it is necessary to not just reduce the output but to actually reduce the carbon dioxide present in atmosphere. Which means that the output should be reduced to a amount less than what can be removed by the natural processes.

What does this mean? Again assuming that the amount that can be removed through natural processes has remained constant****, output must be reduced to pre-Industrial revolution level (when presumably the warming started). There is a problem though, carbon dioxide is just a by-product of energy consumption, as there are no significant non carbon dioxide producing energy sourcing  reducing output in effect means reducing energy consumption to pre-Industrial revolution level, in fact as world population has increased manifold it would imply an unimaginable reduction in per capita energy consumption.

Why unimaginable? Well, energy consumption is a good indicator of the living conditions, generally better living conditions imply higher energy consumption. Therefore a decrease in per capita energy consumption to less than pre-Industrial revolution levels essentially implies living conditions of average human will revert to the same era.

And while many people may hold romantic notion of a pristine earth unsullied by machines, in reality this will mean a life plagued by poverty, hunger, want and disease, a life wasted in daily grind for survival.

Economic holocaust in other words.

I am not sure that global warming is worth it.

*Not really, this was just to bait Respected Ma'am

** Though industrial pollution also means more smoke which I believe reflects lights.

*** Though to put things in perspective, earth has been much warmer than predicted for global warming earlier, and contrary to doomsday scenario for global warming, life thrived. Also extinction of large number of species is a natural phenomenon.

**** Though geoengineering may remove atmospheric carbon dioxide. Its feasibility should be studied.

Wednesday, December 09, 2009

Not that ...

..any of you will have any idea but aren’t restrictions on insider trading kind of stupid?

Sunday, December 06, 2009

The List

Here is what I want to write about if the notion strikes me, swiss minaret ban, Pratap Bhanu Mehta, and Swapan Das Gupta.

Combinations you like

You know what tastes really good ?  Black Currant. Hey wait, Rum n Raisins is wonderful too. A bit of dilemma, isn't it? You know what, go ahead take a double scoop, 'cos you deserve that.

Now 'spose there is a special place in your heart for sci-fi, but wait you also are a sucker for western. Well if that be the case Firefly is just what you need. It is awesome and it is double scoop. Now why the heck was it cancelled!

Wednesday, December 02, 2009

For Code Monkeys

1.       No you don’t have to design before implementation in all cases. You may have read it in some standard software book, or you may have learnt it in your first company, but it is wrong. You should design only if you need to. So when do you need to design? Glad you asked. Here is when you should design. When you think that spending time on design is going to benefit the implementation, this is only when what you are implementing is non-trivial, i.e. some sort of system or framework for system, it also implies that you are familiar with the underlying infrastructure. However if what you are doing in effect is tweaking parameters of some function, and you are not familiar with the underlying implementation, there is no point in having a design.

 

2.       Guess what is not in the ten commandments Moses brought with him from the mountain. If you guessed thou shall comment every line of code, Congratulations. See, comments are for a purpose, specifically to convey information for maintenance of the code. And believe it or not, there is something like too much information. Which is why, comment no more than what is necessary for the intended purpose. This is why comment should not be trivial. Here is what you can comment.

 

a.       A brief comment at the beginning of the function stating the purpose of function and some other critical information. This is useful for automatic documentation.

b.       Inside the function, comment only to provide overview of the algorithm and that too only when the implementation of algorithm is non intuitive.

c.       Similarly comment there is some particular piece of code which is not very intuitive to understand, this may happen when you are fixing some bug.

 

Fortunately rule for what not to comment is simple. Never comment on any thing which is trivial. This means, you need not comment for every damned variable declaration /definition, loop, or control.

 

3.       I forgot, there is one more rule about what not to comment. It is the FREAKING OBSOLETE CODE. Yes, I know some of you are under impression that it is a very good practice. Here is my request, die, preferably painfully. Obsolete code is not documentation and doesn’t provide information in concise form for maintenance. Instead the commented code keeps piling on cycle after cycle and the rotting stench of thousands of lines of obsolete code just repels anyone from actually reading the code. If you are commenting obsolete code for reversion, then either you are too stupid, or you work in a really pathetic company which can not afford any tool for version control.

 

4.       Objected oriented programming is a paradigm which means approaching the problem from a specific direction. Specifically the problem must be understood in terms of objects which are implements as classes and their behaviors which is implemented as functions. This means that if you have flag to track some property of the object it is not required to return the flag in the “Get” function. Instead the proper behavior of the function is to return the property of the Object as accurately as possible.

 

Hindu fascist thought of the day

In secularism Hindu is the Untermensch.

 

Dear (intellectual) wannabes

Not to go all Goldwater on you but, nuance for sake of nuance is NOT a virtue.