The falling American dollar

I read this article about the benefits of a falling dollar. Apart from making the exports more competitive, which accounts for about $ 1 trillion (approx 7-8%) of the economy, it does not add any other major reasons why Americans should not be concerned.
A couple of good points though:
1. Most goods imported from Europe are luxury, so they hit only the higher economic segment. The cheaper goods come from Asia. Will the falling dollar not affect these imports? Are most currencies in Asia linked to the Dollar? And would continued fall of the dollar not drive these investments out of Asia?
2. Prices in the US remain the same....again subject to the point above, but point taken, though this may not remain so.
3. 1985-1990 dollar fell but the economy did well....why? And why should it repeat?
4. China would not cash in on its dollars....completely agree, it would lead to much political instability and hamper Chinese growth, something the Chinese government is counting on for its sustained presence in China.

But the most important point is the loss in credibility of the dollar. The point is that other economies are rising, and investors will be more reluctant to put their money on the American economy instead of other more growing economies.

1) Does the fact that the other economies in the world are doing better and contributing more (Africa, India, China as an example) automatically imply that the strength of the American economy is becoming less important? Does this mean the dollar will continue to remain low?
2) McKinsey came out with a report saying that USA would remain the leading economy for the next 2 decades. Does this mean that other economies (since most are based on some form or other other on the American economy) continue to play second fiddle?
3) How is the money flow for business networks (upstream-downstream) for those completwely excluding the American economy growing? If this reaches a critical momentum, it may marginalize the American economy faster than one expects. That is, if the number of trade cycles (manufacturing, processing etc) between say Europe (value addition), China(manufacturing) and Africa (raw materials) increases, this may undermine the value and strength of the American economy. Is enough attention being paid to these sectors? All it needs is a critical mass.

All said, I hope things work out!! After all, I just got a job in the American economy!!!

OpenOffice.Org

I just managed to download OpenOffice.org on my Vista.....just because I so badly needed a word processing program. And it is not bad. Though I do not know about porting it to Word. I may not need that though....I just converted these pages right into .pdf and .html. And I tried Comic Sans MS (since most non-Macs do not have Lucia Grande....a notable exception being blogger....though I do not know why it has such limited fonts while publishing the blog!!) along with the shadow effect, and though the shadow does not make much of an effect on the pdf, I can now make a decent web page for documentations without having to learn html programming....I think it should make a web page real soon.
In other news, I found enough resources to work on my term paper. My topic will be on self organization of economic systems. I think Schumpeter came out with ideas on evolutionary economics and they are trying to come out with some methodologies for it.....I think!!
Can not wait to get onto Linux....I think there can be very many things to try out there!!
Also, I need to learn some kind of a scripting language like Perl or Python. And on hwo to interface it with the web page. That is, how to use a program to go visit a web page, manipulate the information on it (on your own computer!!) and get data sets out of it. Though this may not be helpful all the time, this along with a scripting language could be powerful.
Here is a small project that might be worth trying:
Build a web crawler to find out the number of web pages containing certain keywords and then run a clustering algorithm on them. I think the field of information retrieval is quite cool! It is just that working with so much data makes the results that come out seem so much better and cooler!! Of course, when things don't work, then making them work also takes a lot more effort.
I think the other day, Google returned one of the blogs I had made private in their search. This angered me quite a lot. When I make a blog private, it should not be visible to the outside world!! There is a reason I am making it private! Another thing about Google that pissed me off was their lack of support to enable a 2 way syncing with the Google calendar. I need some place to store my online journal where it won't turn up on Google searches.....and hope Google does a Google-worthy thing with the calendar. I would really love to use it!! This is another place where learning to manipulate web data using scripts may help!! Go Python....go perl!!
The next 3 weeks are going to be hard....I do not know how I am going to manage....and yet I try to find time to blog!! Either I must be getting really stupid, or it must be the guys at Microsoft, because I could have sworn I was not able to find out how to open a file in the new Windows Media Player.
Things to do before I can breathe again:
1. Finish term paper for CSCS (and return the library book before you have to sell your soul for the fine!!)
2. Clean up concepts on probability.....this includes random processes, more problems on random processes and even more on random processes!!!!!
3. Finish writing the test cases for the smart motor and find out other crappy errors which could never occur.

Tomorrow, I think I will read through the manual and find out all the things which can stop the motor. Then I will need to make the if conditions. I will also need to make the module which reads from the CAN bus. Will have to ask Mike about the module and structure of the message. I think this project gave me a good exposure to a fairly medium size project spread over many files. Now I can try to make the communication code with Domitilla a little better.....more spread out and modular.....will be fun to build the structure of the code. Also, I get a strong feeling it will help understand other open source projects.

It has been 2 hours....its almost 1 in the morning....and I'm still blogging!! Man I waste time!!!
Off I am now....to try study....hopefully!!!

The last 3 weeks as a student....for some time at least....

The final 3 weeks as a student are here....only 3 more weeks to go. And not an easy 3 weeks at that. I just came out of a 5 hour session of solving assignment questions and could not come up (in a combined effort) with half the answers :( Things are always tougher when you have to wrap up projects!! But some things to look forward to......
1. Buy a new hard disk so that I get it just as exams get over....not before so I do not flunk the exams, not too late so I waste time....
2. Install Ubuntu:
a. Move all data into hard disk
b. Install the OS
c. Add Compiz fusion and hope it works on the RAM I have!! Pros: Scale (Expose)
d. Google Desktop for desktop search
e. Some kind of a tool for collecting web data in a decent manner and making a proper journaling system for it
f. Configuring other options to my liking
g. Learning scripting languages like Perl and Python
h. Learn enough to write a web crawler....without having to go through the entire tire of C!!!
3. Read all the left over material that has been piling up
4. Read some of the books I have been dying to read!!
5. Catch up on all news feeds
6. Make an active blog...
7. Get some exercise. A study just said that sitting down automatically stops the burning of fats!! Didn't have the courage to meet the rest of the article!!
8. Get up really really late!!
9. Catch up on all movies and TV serials!!

Like I said.....only 3 more weeks to go for heaven!!! There is also a HDTV in the Commons for me to get my grubby hands on!!!

In pursuit of perfection

My biggest worry is getting everything right in anything! I have tried time and time again to work on it. Like when I get on a new machine, I type the Hello World program, hoping to get it right the first time. Simple program, nothing as complicated as the others I write otherwise.....should be a piece of cake......But then I find the compiler screams out...the problem it seems is something to do with not including the path of the compiler in my $PATH variable. And just like that, I could not even get "Hello World" running on the first try!!
Have a lot of trouble regarding this.....back in school and college it was cool to know the method and the answers but make careless errors. However, slowly the world is getting more and more unforgiving....mistakes are being punished more severely than ever before. The stakes are getting higher. There have been numerous debated on hitting home runs vs singles. In the McKinsey Mind, the author makes a case that hitting singles is far better than home runs (at least if you are at McKinsey) because home runs raise the expectation and then disappoint. Well, looks like I am a home run hitter. And looks like consultancy is not my cup of tea in that case. To me, it looks more like home run hitters belong to the world of venture capital (not entrepreneurship, because you work on only one idea and have to get it right). The problem with that of course is the lack of money :P It is after all what the VC gets to the table :|
The only other options are to look at improving on this elusive skill others call perfection or to remain in an insignificant position where mistakes don't matter. And since the later is not looking like much of a pick up line.....any suggestions on the former?? Anyone??

Consulting

Ace the Case
Cracking the case

Books to Read

What started out as a list of 4 books (some place to keep note so that I could remember when I get back to Umich) has become a list of 52 books I would like to read over the next year or two. This list may change a lot though, but having reached around 30 or odd books, I came to the thought that if I had only a year to live, what would I do? Ideally I would like to travel around the world and read these books. I would certainly want to do this before doing my MBA, the moment immigration allows! Anyway, the list of books :

1. Social Intelligence
2. Emotional Intelligence
3. Fabric of Reality
4. Singularity
5. Godel, Escher and Bach
6. Guns, Germs and Steel
7. A new Kind of Science
8.The Art of Computer Programming (Vol !, II)
9.Mckinsey Mind
10.McKinsey way of thinking
11. Cat's Cradle
12. Doors of Perception
13. Midnight's children
14. Quest for consciousness
15. Society of the Mind
16. General Introdcution to Psychoanalysis
17. Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution, and Epistemology, Gregory Bateson
18. John von Neumann and Oskar Morgenstern, Theory of Games and Economic Behavior (1944)
19. Norbert Weiner, Cybernetics (1948)
20. Richard Feynman, QED (1985)
21. Surely You are joking Mr. Feynman
22. The Mind's I
23. The Origin of Species
24. One, Two Three, Infinity
25. Atlas Shrugged
26. Fountainhead
27. The Picture of Dorian Gray
28. My Life
29. The Audacity of Hope
30. Freakonomics
31. Palestine Peace not Apartheid
32. State of Denial
33. 1984
34. Catch-22
35. Brave New World
36.The Grapes of Wrath
37. Catcher in the Rye
38. Slaughterhouse 5
39. A passage to India
40.Of Human Bondage
41.A Farewell to Arms
42.Call of the Wild
43.Kim
44.Satanic Verses
45.Development as Freedom
46. Economics : Samuleson
47. A Wealth of Nations
48. Animal Farm
49. India Unbound
50. Reasons of State
51. The Art of War
52. The Future of Freedom

A List of 100 books that shaped a century of science

A List of super hot papers published in science

Pandora

Pandora is a music exploration software. I also think it is a brilliant idea. And though my friends tell me it isn't the first, I was completely blown away by it! However, I felt like most things it could be improved. (Remember, now inspired by TED, I am now dreaming!!) So I sent them a mail. Feedback? Here it is :

Hi

I just came to know about Pandora in the past few weeks, and I think this is a great product! I have discovered more songs I liked in the last few weeks that I did in a couple of months before using Pandora.

However, I had a suggestion, though it could prove to be hard to implement. Rather than just make a selection based on a song, why not make the selection based on a part of the song? The algorithm seems to focus on attributes of the song like male vocal voice and electric guitar, so you may have a lot of the information needed to implement this.

Take for example, Metallica : Master of Puppets. Maybe the listener does not like the whole song, but he/she "really" likes the electric guitar solo piece. He/she should be able to modify the search algorithm to include only the information about the electric guitar and not the vocals. Then, say it would make more sense to play Ecstasy of Gold rather than November Rain. They would give very different search results and the latter would be a more effective search.

Another thing that could be done is to give the user an option (turned off be default except for the first time he registers) to choose which of the options he liked in the song. For example, if we had the vocals and the guitar at the same time, he/she may choose only one of them, whichever one he likes, or both. There could be an option in the menu which enables him to choose whichever component of whichever part he likes.

I believe both of these features will enable a better search. What we have right now is not the DNA, but the whole cauldron of DNA. Making the search even better would really let us choose the DNA we like without getting a baggage every time we choose to like something.

As I said earlier, you guys are doing a great job!! Please do let me know what you think of these ideas.

Also, is there any reason we can only say whether we like the song or not but not rate how much we like the song? Is that related to copyright? Would such a rating not make the search many times better? And you must already have all the information related to this. Just that you would have to input the ratings as weights and implement these in the search algorithms as weights. Of course I could be wrong, but this should not be terribly difficult, should it?

Thank you for taking the patience to read the email.