Some networking / No drugs - the market has failed / Clever novelists / Business…
Humiliating to human pride as it may be, we must recognize that the advance and even the preservation of civilization are dependent upon a maximum of opportunity for accidents to happen. F.A. Hayek
The current market system does not guarantee sufficient supply of new drugs. It is simply not economically sensible to try to find new antibiotics, since it takes a lot of money and the sales will not be as high as with Viagre, Lipitor, etc. The other reason has been the price regulation by the US government. It has simply not paid to produce flu vaccines. So, now there is a shortage of flu vaccine. But this is bigger problem. Should the states somehow finance development of new, essential drugs? Like new antibiotics, new vaccines (e.g. AIDS-vaccine, which does not make economical sense to develop - the AIDS drugs are a better and continuos business - a vaccine would only be a temporary business). I thin they should. They could do it by promising to pay (collectively, via U.N or WHO) certain billion of dollars for the company, which develops the next antibiotics. By doing so, there would be economic incentive for the R&D and there would constantly be enough new antibiotics in the pipeline. The U.N or WHO would then own IP for the new drugs and license production to drug manufacturers. This kind of system could work. I know it has been proposed earlier, but I just wanted to rant a little.
I just received Business Miscellany from The Economist. It contains all kinds of interesting, if not useful, statistics about businesses. For example, that BASF stands for Badische Anilin und Soda Fabrik, that average French person works for 1453 hours per year, a Finn 1713 hours, and a American 1792 hours. But in Norway they only work for 1337 hours per year. Having oil does not hurt. The highest salaries in 2003 were in Germany, not in USA. And in 2003 euro was not as strong as it is now. So, those who want to make a lot of money shoudl apply for jobs in Germany, not in California. There are 719 billion dollars worth of dollar notes in circulation, but only 186 billion euros worth of euro notes. And so on, a very entertaining book. And it was free, gratis!
Now I will leave for Watertown and Mt Auburn cemetery on my bike, the weather and foliage are too nice for spending whole day here inside. Not even thinking of complexity is better than cycling in autumn...
It was a good rire. I took many photos of the River Charles Waterfall, but I did not make it to the cemetary. Good anyway. I also had a lot of time to think about complexity, and the general structure of my thesis.
One can only fish 25 herrings per day by the waterfall. And one must not fish any on Tuesdays, Thursdays, or Sundays. The herrings have other things to do than entertain fishermen those days.
I am getting seriously interested in my thesis topic. I am sure that there is much new, and quite easy, research for me to do, and that doing it will be very blissful. At least if things are going as well and enjoyably as they did on Monday. I also think that about 1 month is just right time to devote for a SDM-thesis. Not much more anyway.
The way I now think about complexity is more or less this. Let us have two systems M' and N'. Then, have a coding function c of systems using a language L. Let's define c(M') = M and c(N') = N. We call M the model of M'. Let's us then denote l(M) the length of the minimum program t(M) in some Turing machine T, which produces M in L, and l(N) similarly. Now, if l(M) > l(N), then K(M) > K(N), where K(M) is the complexity of M.
Since a system does not contain any information, we cannot compare two systems in terms of complexity. But we can compare their codings, and we can use the above definitions (or something similar, there may be some errors still) to order systems in terms of the complexity of their codings in L. A practical problem remains, though. At least in my thesis, I have no way of defining the methods of getting from c(M') to l(M). Thus, I have to use some proxy methods. This is actually what all other papers are doing, but no paper I have seen so far gives any theoretical reasoning for the approach used. I gave here one possible theoretical reasoning. I assume this is what writing a thesis is about.
Of course, in my thesis I will write the argument above in much more detailed manner. I omitted several important steps, definitions, and assumptions above. But I am happy with the progress I made on Monday.
Alun perin julkaistu kuukausikoosteessa "Blogs November 2005".
Some networking
Some Sloan club arranged a networking happening in Hotel@MIT. Networking happenings are strange. One goes to some place, has some wine, some canapes, and discusses with others who happen to be there. I talked to one older manager form the World Bank, a manager from UN, and some people from energy industry. Interesting. Maybe UN would be a nice place to work. I sent them an application (using their websystem), but they have effectively ignored me so far.No drugs - the market has failed
The current market system does not guarantee sufficient supply of new drugs. It is simply not economically sensible to try to find new antibiotics, since it takes a lot of money and the sales will not be as high as with Viagre, Lipitor, etc. The other reason has been the price regulation by the US government. It has simply not paid to produce flu vaccines. So, now there is a shortage of flu vaccine. But this is bigger problem. Should the states somehow finance development of new, essential drugs? Like new antibiotics, new vaccines (e.g. AIDS-vaccine, which does not make economical sense to develop - the AIDS drugs are a better and continuos business - a vaccine would only be a temporary business). I thin they should. They could do it by promising to pay (collectively, via U.N or WHO) certain billion of dollars for the company, which develops the next antibiotics. By doing so, there would be economic incentive for the R&D and there would constantly be enough new antibiotics in the pipeline. The U.N or WHO would then own IP for the new drugs and license production to drug manufacturers. This kind of system could work. I know it has been proposed earlier, but I just wanted to rant a little. Clever novelists
The big booksellers order new books based on the sales of the previous book by the same author. If the sales were not good, they will not order many copies. This has forced authors (novelists especially) to assume pseudonyms, when sales of one novel have not been too good. Today's Wall Street Journal had an interesting article about this.Business Miscellany
I just received Business Miscellany from The Economist. It contains all kinds of interesting, if not useful, statistics about businesses. For example, that BASF stands for Badische Anilin und Soda Fabrik, that average French person works for 1453 hours per year, a Finn 1713 hours, and a American 1792 hours. But in Norway they only work for 1337 hours per year. Having oil does not hurt. The highest salaries in 2003 were in Germany, not in USA. And in 2003 euro was not as strong as it is now. So, those who want to make a lot of money shoudl apply for jobs in Germany, not in California. There are 719 billion dollars worth of dollar notes in circulation, but only 186 billion euros worth of euro notes. And so on, a very entertaining book. And it was free, gratis! Cycling and thinking
Now I will leave for Watertown and Mt Auburn cemetery on my bike, the weather and foliage are too nice for spending whole day here inside. Not even thinking of complexity is better than cycling in autumn...It was a good rire. I took many photos of the River Charles Waterfall, but I did not make it to the cemetary. Good anyway. I also had a lot of time to think about complexity, and the general structure of my thesis.
One can only fish 25 herrings per day by the waterfall. And one must not fish any on Tuesdays, Thursdays, or Sundays. The herrings have other things to do than entertain fishermen those days.
Some clarity on complexity
I am getting seriously interested in my thesis topic. I am sure that there is much new, and quite easy, research for me to do, and that doing it will be very blissful. At least if things are going as well and enjoyably as they did on Monday. I also think that about 1 month is just right time to devote for a SDM-thesis. Not much more anyway. The way I now think about complexity is more or less this. Let us have two systems M' and N'. Then, have a coding function c of systems using a language L. Let's define c(M') = M and c(N') = N. We call M the model of M'. Let's us then denote l(M) the length of the minimum program t(M) in some Turing machine T, which produces M in L, and l(N) similarly. Now, if l(M) > l(N), then K(M) > K(N), where K(M) is the complexity of M.
Since a system does not contain any information, we cannot compare two systems in terms of complexity. But we can compare their codings, and we can use the above definitions (or something similar, there may be some errors still) to order systems in terms of the complexity of their codings in L. A practical problem remains, though. At least in my thesis, I have no way of defining the methods of getting from c(M') to l(M). Thus, I have to use some proxy methods. This is actually what all other papers are doing, but no paper I have seen so far gives any theoretical reasoning for the approach used. I gave here one possible theoretical reasoning. I assume this is what writing a thesis is about.
Of course, in my thesis I will write the argument above in much more detailed manner. I omitted several important steps, definitions, and assumptions above. But I am happy with the progress I made on Monday.
Alun perin julkaistu kuukausikoosteessa "Blogs November 2005".
Kommentit