Douwe Osinga's Blog: June 2003

Monday, June 30, 2003

The jailhouse rock

There is a category of social problems for which current solutions just don't work and still we don't really seem to care. We really want these problems to be solved and we for some reasons the solution strategies we have are the only ones we can think of. Not trying these strategies is like not acknowledging the problem. Still, it's silly. Some examples are: development aid (doesn't work, but we'd like to do something), secondary school teaching (most people forget everything they learn there) and punishing criminals by prison terms. About the first two maybe some other time more. Now for prison.


Crime punishment has of course a number of functions. Revenge, keeping bad people out of society and so on, but the most important one is to make sure the criminal won't do it again. Originally, re-education was a big item here, but over the years at least in the US, the plan seems to be to deter criminals from doing it again by punishing harder. Either way doesn't work. If criminals are educated in prison, then it is only in new ways of crime and if they come out with plans to better their lives, it is mostly the plan not to get caught again. As for deterrents: if hanging somebody for stealing bread in the nineteenth century didn't work, what is? Take a look at the statistics. In America, 5.1% of the population will serve time at some point in their lives. For black males, this is 28%. 67.5% of released convicts gets arrested for something serious within three years of release (on average more then three times).


So why do we keep putting people in prison? I don't really know. Partly it has to do with the revenge part. For example: say we have a murderer and you are the judge. You can give him life, or this new pill, which alters his brain state and makes sure he'll never murder again. The catch? The pill does not work if combined with a prison term. You choose. Maybe the more open-minded readers of this blog will tend to choose the pill, but the general population? Barbertje moet hangen.


There is the more cynical view that prisons mainly function as a marking and sorting system. Our complex society tests its members, just like products in a factory. Those that are unfit are thrown in prison, not to punish them, but to mark them as ex-cons in order to make sure that they will not be hired by accident. If you think there's something to this, read Michel Foucault.


There are two alternatives I can think of, that we should consider. I'll present one here now and one other somewhere in the future. The first alternative to prison as a punishment is torture. Now I realize that torture has a bad name, but I think this is more to associations with the Spanish Inquisition than to fact. Torture as punishment should be clean, medically supervised and not scar, should not lead to any lasting physical problems. The only thing is that it should hurt. Hurt badly. So badly that people won't do it again, or even better, won't consider doing it in the first place.
Before you decide that torture is inhumane, think about the inhumanness of prison terms. If somebody walks out of prison after ten or twenty years, you have effectively not only destroyed the last years of his life, but also the coming years; it is very hard to get back into society. If you torture somebody and this lasts only one day, nothing but punishment has been applied. The person can go to his job the next day as if nothing happens. He gets a new chance.

I'm not a roll

Two weeks ago, I opened my blog with: I'm not a roll, and while I'm really not, I meant that I was on a roll, because I had blogged three days in a row. I'm a messy speller and an untidy typer, something the more regular reader of these pages will not have escaped.


A week or so ago, I came across this article by the New York Times, about a weird machine that supposedly improves your ability to draw dogs, spot prime numbers and proof read. It works by suppressing certain kind of brainwaves making your brain more like that of an savant. What happens according to prof… is that normally our brain is in abstract mode, which makes it harder to do these more literal tasks. We know what a dog looks like, but to draw it, to literally transfer what we see to the paper is hard, because our abstract thought machinery comes in between. The same goes for proofreading (I know how words should be spelled, but anything close to it is good enough and the me part of my brain only gets the interpreted version of what I read, so I never see an spelling error) and for spotting prime numbers. Apparently our brains are pretty good at integer calculations at a low level; at a higher level this ability is obscured.


You might go one step further. There are the more abstractly thinking people, who don't see what they see or think, but only see the abstractions of what they see or think and the more literally thinking people or see what they see and think what they think. The first way of thinking is maybe more found in people who study science, the second way of thinking is more common to those who study arts or languages.
Both ways of thinking have their applications of course. Also, this only has to do with how people approach problems in their thinking, not much with how creative people are. If this theory is true, you probably would find most artists in the second group, because the ability to copy what you see or think in a literal way into what you draw or sculpt, obviously helps in being acknowledges as an artist. The creative people of the first group would tend to end up as scientists, inventors or sometimes writers and probably the more sober writing writers (complex scenery descriptions are like paintings). I do think the world is missing out on artists of the first group. Maybe blogging can grow out to become an art form for this group. (Yes, that's me.)


Another funny thing is the relation towards computers. The second group usually doesn't like computers much. They can me made to see that computers are useful or even essential to some endeavours, but there is no love there and hardly ever the overwhelming feeling of wow that people from the first group tend to experience. It's easy to see why. The things computers are good at, are also the things that people from the second group are good at. It's more a thread than help. For first-groupers computers are a much better match. The computers fix spelling errors (for example, I wanted to type spelling with triple l, but the auto correct feature of word wouldn't let me), can do image processing and calculate any number.

Thursday, June 26, 2003

The Matrix Reloaded: beyond meaning

So I went to see the Matrix Reloaded yesterday. In a way I think the movie reached a new level, a level we could have seen coming since the first James Bond movie. In the earliest movies, fights were pretty real. Heroes got hit and hurt. But that couldn't last, because a hero is more. So heroes got hit and should have hurt badly, but didn't because they were heroes. The villains of course, had to keep up; otherwise the movies wouldn't be very attractive. The same thing happened with car chases and other Hollywood action up to a point where the link between action scenes and reality became less and less strong. Fight scene become more like dances as do car and helicopter chases.


In a way, this happened to other parts of the movie too. Everything becomes less likely, because the more likely has been done before, up to a point where the viewer no longer cares about the likeliness of a scene and the scene (hopefully) only appreciate for the visual beauty of it. The Matrix Reloaded is the next level in that the Wachowski's brothers finally have let go of the requirement that the story as such should tell a story. The Matrix Reloaded is series of nicely done scenes with lots of special effects, exploding attributes and nicely choreographed fighting, mixed with mysterious scenes where people like the Architect, The Oracle, The Keymaker and even Morpheus hold little speeches seemingly making some philosophical point.


These mysterious scenes are just like the fight scenes, there is no longer any link with reality or even philosophy. It's the sounds of the words that count, no longer the meaning. What can be said about philosophy in movies, has been said before. It is more like poetry then philosophy. The Matrix Reloaded is beyond meaning. They say its part two and we're waiting for part three, but there is no real reason to assume that this was part two and not part three. It is beauty and action for its own sake, not for the sake of the story or some Idea, because there is not Idea, there is no story.


I had fun though.

Sunday, June 22, 2003

Software with less bugs

According to the NIST about 80% of the time of software developers is spent on finding and fixing bugs. Not that all bugs are found. Think about it. This means that you could write the program five times over if you didn't mind that it had errors. That got me thinking.


What if we would design the program, decide which subroutines we want and then write the program three times. We can then run the three programs and compare the results. If the results of a subroutine are the same, everything is ok. If they are different, probably two of the routines give one answer, then the one with the different answer probably is faulty. We don't have to find the error, we could just take one of the other two. No need to write test sets, or scanning the code for what goes wrong.

Saturday, June 21, 2003

Making shopping easier

Men generally don't like to shop for clothes. Men usually have their favourite shirt, favourite jeans and a favourite sweater and they are good. Shoe shopping, for some women an activity bordering on the therapeutically, is for men nothing else then shoe replacing. It is like shopping for milk. The old milk carton is empty and needs replacing, just as the shoes need replacing because of a hole. There is no point in trying to find a better or nicer milk carton, the usual will do, thank you very much. For shoes and clothing in general, this doesn't work. The old stuff is never available and the whole selection procedure starts over each time I want to buy shoes.


What we should have (and probably do) is a unique code for each piece of attire. You buy the stuff, note the code and wear it. Then when your favourite jeans have a hole, you go to this website, type in the code and voila, they'll send you another one of your favourite pair of jeans.


For groceries this scheme could be simplified. I was thinking about a bar code scanner next to the garbage can. If you throw the package of something away, you think, do I need this and this replaced and if yes, you scan the barcode. A computer will then automatically order a new carton of juice or whatever it is that you need. Alternatively, it could just add an item to your shopping list, which would be more feasible, but far less cool.

Friday, June 20, 2003

Patens and medicine

Do patents work? They seemed to have worked in the nineteenth for getting all those nice technical inventions out there. Pharmaceutical companies use them a lot and so do software companies and biological companies now. It always has been a bit strange: patents are supposed to create innovation by allowing monopolies, not really institutions known for innovation.

Copyright, patents and brands have been used to make the information society possible, that is, to make the production of knowledge much like the production of things. Of course the production of knowledge is not like the production of things at all. Information can be copied without cost. Things always have the monopolistic property that they can have only one owner.  Information as they say wants to be free.
Innovation however, sometimes needs motivation beyond curiosity.  Medical research costs a lot of money, as does some industrial research. Without the protection of patents nobody would undertake this kind of endeavours, or so the reasoning goes. Without copyright protection, writers wouldn't write, composers wouldn't compose etc, etc. The last bit is not true. Writers have written, composers composed and painters painted since the dawn of time without any kind of protection. I'm writing this because I like to, not because the copyright protection. But you can't really build an industry around it and as our society becomes more and more an information society, this becomes a problem.

Let's focus on medical research here for a bit. Patents allow medical companies to make their money back, the money they spend on medical research. This works, but is not a very efficient way of doing things from society's point of view.

Have a look at the graph to the left. A company has produced a new drug and is selling this drug at monopoly price p1. The line is the price/demand relation. People that can afford p1 effectively have to pay back the research costs. This results in a demand of q1 and gives us our company revenue of q1*p1 is the striped area. For society as a whole the deal is that it has to pay q1*p1 each year to pay back the research done for the medicine.  Clearly there is enough demand for it. In a market environment, however, this leads to the situation where more then half of the people do not get the medicine, even if it could be produced against little extra cost. Indeed this is happening with Aids.

If we would somehow succeed in getting in a way that the price depended on the income of the people, everybody would win. The people that can afford prices>p1 will have to pay less then before, because the people that could not afford the medicine will now pay something of the p1*q1 amount of money needed. The people that couldn't afford p1 before obviously win, because now they get the medicine for a price they can afford.

Manipulating the market for better outcomes tends not to work, but the example above shows that the current situation is sub-optimal. Some other day, I'll post about a solution I thought of.

Tuesday, June 17, 2003

Making the world funnier

The simple reason why they play laughter in comedies, is that the jokes seem funnier when more people laugh. The viewers might think the taped hilarity irritating, but he or she enjoys the show more. This got me thinking, why don't have laughter boxes in every day life situations. The jokes we tell each other are not that much worse then the stuff that passes as prime time television, it is the canned laughter that makes most of the difference. So, if we outfit every office with a big red button that chuckles as opposed to the mythical Internet button that does nothing, we would all laugh more and enjoy each other more.
Another reason why people in comedies seem funnier that people in real life, is that they have text writers. That seems harder to implement in day-to-day life, but it is possible. I mean, situational comedy is based on the situation one is in and we mostly share the situation we're in with a lot of other people. Who cares if a hundred or even a thousand other people crack the same jokes as I do in the same situation that I am in, what is important is that the people I tell the jokes to don't know. Really, even that is not important, important is that they don't know the jokes and can laugh about them. So, if we pipe the jokes through mobile phones, e-mail or whatever is your favourite modern communication medium, then the joy gets spread around more.
If everybody had a script, we could even work out complete plays. If the lines are piped in on a last minute base, it would be a truly mixture between watching a set piece and playing in one. It would be fun to be in one, because you would lead the life of somebody else while completely experiencing that life. And it could be executed parallel with whole groups, i.e. we could have an episode of say Friends playing a hundred times parallel with 600 friends. Could be fun.

Sunday, June 15, 2003

Atlantis II

I just added a project sealevel with some code that lets you view the map of Europe at different levels of the sea. Nice to play with.

American Anxiety

A couple of days ago I watched ‘Bowling for Columbine', a documentary about gun possession and why the murder rate is so high in the United States. The documentary did actually not conclude that it had to do with a high rate of gun ownership. I did some research and here are some facts to back that up:



For whatever reason, it seems that Americans are just more murderous than Europeans, Japanese or Canadians.


Another striking difference between the United States and Europe is the number of hours worked per person. In the Netherlands, the average number of hours worked in a year is about 1400. In the US it is 2000. Americans are richer then the Dutch, but that can be contributed almost entirely to the difference in hours worked, i.e. Americans and Dutch earn about the same per hour, the Americans just work more hours.


Now, in the Netherlands, as people have become richer, they have also started to work less hours, which seems logical to me. The more money you have, the less important it is to get even more, so the value of free time relative to money increases. In South Korea the number of hours worked is something like 2600 a year, but it is dropping. In America the same process was going on, until the seventies or so, when suddenly people started to work more. Indeed, a lot of the economical growth was fuelled by just more work.


Why would Americans work so much? Michael Moore seems to indicate that he thinks that the high murder rate has to do with anxiety in American society. While I'm not sure that that is a reason, he makes some good points that anxiety is indeed more of a problem in America than it seems here. Could it be that that is the reason why Americans work so much? Because they are afraid that they would end up poor and have nothing, or will not have enough money to put their kids through college or that they will loose their jobs? On the other hand, that would probably lead to a higher savings rate and that is not going on.

Wednesday, June 11, 2003

The location of Atlantis

I'm on a roll - three postings in three days! The subject for today is atlantis and where it might have been. A lot of theories and the most likely theorie probably is that it never existed, but I have another, one I haven't seen published, but let me know if I'm wrong.

Atlantis is the Mediterranean. 10.000 years ago, around the time Plato places it, the world was just coming out of an Ice age. Sea levels were rising. Since the sea was so low that the street of Gibraltar was closed, sea levels in the Mediterenean where even lower, because the Mediterenean currently needs water from the Atlantic to keep up its current level. One day the street of Gibraltar opens and you have the sort of disaster Hollywood would love.

Now look at the map above. This is what the coast line would look like without Gibraltar blocking the Med. If it would, sea levels would drop. The pink area's would also become land.  That is where atlantis was. When the floods struck, civilisation is thrown back some thousand years and the people are left with some vague memories about great floods.

Tuesday, June 10, 2003

Not a children's bible

This weekend I was reading a bit in a children's bible and as always it struck me how far removed the childrens stories are from the cruel reality of bible tales. Childrens bibles are very selectively edited (of course).


I was thinking it might be fun to create a adult bible. A childrens bible only contains the stories that are suitable for children; my bible would only contain stories that are unsuitable for children, or for that matter most Christians. Because the bible is a book full of strange and weird stories, mostly wasted on and ignored by mainstream Christianity.


The storie of how the sons of God had intercourse with the woman of the earth and that this union produced the giants, men of renown is mostly missed, is one of my favourites. Then there are the more bloody stories of how the Israelites tried to get rid of the tribe Benjamin by genocide or the killing of 42 little children by God because they laughed at a balding profet. The Onan story of course is one the weirdest.


I really think it could be a best seller.

Monday, June 9, 2003

Why languages are complex

Why do have languages all these complex grammatical features? Languages seem to become more simple over the ages, see Lating becoming Italian, Sanskrit becoming Hindia and Old Greek becoming New Greek for example. But if this is a language law, how did we end up with the complexities in the first place, or is this rule in place only for the last two thousand years. If you think about, people have been speaking languages for hundreds of thousands of years. Until two thousands years ago, this resulted obviously in rather complex languages (asuming they started out simple). And then suddenly they started losing complexitiy.


I think it might have to do with the fact that complexity in languages is useful before the discovery of writing. Spoken language in pre-writing culture is used not only for communication, but also for remembering complex stuff, epos's like the Manas with more then 500.000. lines, or even mathematics in de Veda with systems to remember numbers. The more complex languages are, the more options the poet has for encoding things in a way that can be remembered, i.e. if a language is completely free of complex rules and you have half a sentence, you have hardly any way of recovering the other half. In a complex language you do. The complexity functions as a form of conistency checking. You loose some bandwidth, but you gain in situations where the reception isn't too good.

Saturday, June 7, 2003

Why governments always need to cut spending

The Netherlands have a new government. Unfortunately the economy is bad and they need to cut expenditure. It seems logical, but if you think about it for a minute, it is strange. I mean, the economy is bad, it is hardly growing, but that means that in real terms about the same amount of money is available for the government as last year, i.e. in money it is actually more. Why does the government have to cut expenditure when they have the same amount of money as last year? This can only mean that they automatically increase spending each year in real terms, i.e. not only do they increase spending by as much as the inflation, but they also add something each year on top of that. And not to do new stuff, just to do the old stuff. Government gets more expensive each year with quite a bit more then inflation. Apparently we need to cut spending just to spend the same amount as we did before. The beast is hungry indeed.


But what about the government wages? Isn't is logical that the real spending of government increases each year for the same amount of government because wages of civil servants increase? Maybe. But that would mean that the productivity of these civil servants does not increase. If productivity of civil servants would rise in line with the productivity of the market sector, then the wages would go up by about the same amount as the productivity and we would just need less civil servants for the same amount of government. Clearly this isn't happening in reality. On the other hand, if productivity does not increase, one can doubt whether wages should.


One way or the other, it seems best to me, if we would decide what it is that we want the government to produce, see how much that cost and treat any price increases above inflation and failure of the politicians involved.

Tuesday, June 3, 2003

Google Share

I stumbled on a nice graph component and couldn't resist. GoogleShare, calculating the mind share of something, is a new project. Enjoy.