Annual millions of dollar, euro, yen (¥) and other currencies are wasted on the fight against smoking. Enormous expenses with only modest successes. The same can be said for the AI - research. Of course, such disproportion has its courses. We should go into the problems of the "artificial intelligence" (AI).
In (8) intelligence is defined as follows:
In order to avoid misunderstandings is valid until further notice: The bases for intelligence and creativity are to comprehend, to remenber and to retain. I think to solve a problem, how it is defined above, is the real intelligence based on convergent thinking. The term creativity is further valid in the above defined form and is based on divergent thinking.
Fundamentally human ways of thinking can be classified into two groups : the convergent and divergent thinking. The question about the usability of a paper clip is a widespread test in order to determine the ways of thinking. Very convergent thinkers cling to one paper clip and hardly suggest new usabilities for it. In most cases this is a kind of clinging. Very divergent thinkers immedially enjoy the thought of creating unusual things with several paper clips. Nevertheless the number is not so important. The different kinds of combinations between things and effects are more important. We will come back to it. The previous experiments to simulate human thinking are exclusively based on convergent thinking. As we do not know now as before how does the convergent thinking work, the difficulties in model it are considerable. At first one example of convergent thinking, the translation of any text into another language. Certain successes in the translation of special texts are inconsiderable. The translation of standard formulations is less a question about intelligence than more a question about storage capacity. What did my university teacher say? "Understanding means recognition." As we are reminscing, one further example : During my long - standing study abroad at the school of professor Polovinkin the most difficult language problem was not to understand the lessons rather than the change from "balls" to "gramme" in the ice - cream parlour.
Especially children and highly - creative people are able to think divergently. In the course of time the convergent thinking will become determined, which is conditioned by the enviroment. Therefore the convergent thinking prevents new ideas. Of course, this process can be returned, too. But at first an example of divergent thinking. Glas ampullae with heat - sensitive serum are melted by gas - burners. If the flame is very hot the melting process goes fast, but the serum can be spoiled. If the flame is weak, the melting process will go slowly or little capillaries will be left in the glass. How would you solve this problem? Do think it over! Mostly the convergent thinker suggests to test whether the ampullae are closed hermetically. If you are a bit intelligent, you can get the idea to find such temperatures experimentally, where the loss is least. The characteristic feature is, that they cling to a weak flame. Certainly, it is possible that convergent thinkers have a good idea, too. Divergent thinkers intensively try to solve the real problem. This is not the hot flame but the heat sensitivity of the serum. But I do not want to show too much. Apparently it is more difficult to simulate divergent thinking. However this is a fallacy. Big parts of divergent thinking have been algorithmized by long - standing research. This is concretely realized by the computer - aided - inventing.
The position of the computer - aided - inventing which is explained here is based on the works of two opposing schools. I think the most important school is the school of professor Altschuller and he is a symbol. As a young engineer and inventor he began to deal with the algorithmization of divergent thinking around 1946. In contrast to most other specialists he already was a successful inventor at that time. In the beginning of the sixties he had the following methodical instruments :
The ARIS in its modern version is only used for the solution of problems, which are not yet can be solved with the standards. As the ARIS is extremely universal, at this place no example, but only a few short cuttings : (very much simplified according to 5)
Ideal machine : The perfect, that is the ideal machine weighs nothing, takes no space, consumes no energy, produces no refuse but nevertheless it fullfills the same work for which it was created. Besides the ideal machine works without interruptions and times of standstill. (1)
administrative contradictions : This contradiction is on the surface of each inventive task. Something has to be done but nobody knows how to do it. This is an almost insurmountable hindrance for convergent thinkers.
technical contradictions : Finally every administrative contradiction underlies technical contradictions. If one part of the technical system was improved with conventional methods, another part would get worse automatically. The designer normally does not solve this contradiction but makes a compromise. The contradiction can only be solved with the help of an invention.
method principles : Here are meant the solution methods for the 1482 contradictions, which are known up - to - now and possible theoretically. They were gained as a result of the analysis of leading patent specifications. Combinations of several method principles are especially effective. Although this method is not undisputed, mostly it shows the approximate direction of solution.
Example : A cargo ship is subdivided into homogenous sections. The ship is able to be made longer or shorter if required. (push boat)
...Tables of solutions : They give information on how to use the method principles for the solution of one concrete contradiction.
Example : Mine railway - it is too light to tow heavy carriages. The wheels spin easily. But the locomotive has to be light to move it deeper and deeper.
Technical contradiction : mass of the movable object <---> power (transfer)
In the sixties professor Altschuller began to transform his methodology to mainframes. As the programms were not able to solve problems and the inventors had no access to mainframes this attempt failed.
However he had achieved a breakthroug in the usage of his theory because of the foundation of two universities of his own and with branches in the military - industrial - complex. But he hardly was supported by state - offices. His main support were the Union of Inventors and the Engineering Union. These powerful unions made it possible for the Altschuller school to supervise the most important developments on all special fields and to prepare it methodologically in order to make it available to the research.
He pushed forward into securityrelevant areas. In this case his "substance - field theory" was very important. This theory made it possible to discuss the solution procedures but the actual problem remains unknown.
Since that time the methodology has been completed with the following instruments :TRIZ : the "theory for the solution of inventive tasks"
Example : The aerial of a radio telescope has to be sheltered from lightning - stroke. For this reason lightning - rods from metal were erected around the aerial. Unfortunately theese bars reflect the radio waves. It is not allowed to fix lightning - rods even to the aerial. What has to be done?
Transition to the mini - task : The positive quality, the shelter from lightning, is conserved. In contrast to this the negative quality, the radio shadow, disappears.
Administrative contradiction : We have to do something but we do not know how to do it. The aerial of the radio telescope has to be sheltered from lightning - stroke, but it is not clear how can we do it without occuring of radio shadows.
Technical contradiction : If you have improved something with conventional means, another thing would get worse. What will be changed with the solution of the task? The harmful factors, which are influence the object from the outside (lightning - stroke). What will be changed thereupon inadmissibly? Visibility, loss of information, reliability and accuracy in measurement.
Solutions according to the table : Principle of seperation (most frequent naming). The disturbing part is to seperate from the object.
Physical contradiction : The lightning - rod has to be conductive in order to shunt off the lightning, but has to be non - conductive in order not to cast a radio shadow.
Standard solution : If an object should be in various physical states periodically, the object has to make the transition of itself on the way of transformations, which are physically reversible, for example, transformations, ionisation - recombination, dissociation - association etc., see data bank of physical effects. It is a question of a fundamental technical law.
Solution : "Lightning - arrester with lightning - rod in form of a gas discharge tube. If a ligthning develops, the lightning - rod will switch on "itself" : The gas will be ionized and thus conductive. After the lightning the ions recombinate again, the gas become electrically neutral, the lightning - rod has no longer electrical conductivity and therefore does not produce a radio shadow." (SU 117 497)
physical contradictions : Contrary demands are made on one part of the technical system.
Example : It must exist a substance in order to cause a determined effect, but its existence is not allowed because it is, for example, food.
substance - field - systems : Professor Altschullerïs opinion is, that an abstract technical system consists of a workpiece , an instrument and a field of any nature.
Fel
/ \
S1 -- S2
Fel - electrical field S1 - luminous substance S2 - electrons
In reality S1 and S2 are technical systems as well. Theese are the display mask and the electron gun. The concrete structur of the system depends on the task, which is to investigate. In this case I consider the television from the view of picture element generation.
laws of the development of technical systems : The development of technical systems obeys the common systemtheory. In order to illustrate this here an example : (1)
The development of technical systems goes into the direction of its perfection, i.e. increase in the degree of ideality; goes with transition to the higher system; etc.
standards : Theese are formalized solutions of concrete inventive tasks (task classes). (2)
Example : If a field can be controlled badly (demonstrate, compare, change or transform into another field) and it is required to guarantee an effective control. So is valid :
a - transform the given field F1 with the help of a transform substance or two substances, which are influence each other reciprocally :
F1' / F1 => S \ F1''
b - introduction of a substance S, which changes its quality because of the influence of the field F1. This change can be easily controlled with the help of a field F2 taking effects on S.
F2' / F1 => F1 --> S \ F2''
Now let us have a look at the following school. Professor Polovinkin began his work on the field of data banks of physical effects as well as on the field of theoretical bases of the computer - aided - inventing. At the end of the seventies he intensively proceeds to the problems of morphological tables. After the beginning with the project of the "fifth computer - generation" in 1981, the government subsidies for Polovinkin became more and more. As they were not able to compete from the hardware - side, they decided to develop the existing advantage on the field of computer - aided - inventing and thus find a more effective solution with less expense. Thereupon the work of professor Polovinkin was expanded with the co - operation of many universities, colleges and institutes of the Academy of Sciences. The project name was "engineer". Now the aim was to develop the existing CAD - systems in that way, that problems arising while the construction were able to be solved by the computer on the level of an invention.
Although he has been oppossed the school of professor Altschuller at all times, Polovinkin assimilates it more and more. At present differences still exist on the following fields :
substance - field - analysis : Professor Polovinkin considers a technical system as series of physical effects. This is very difficult for average intelligent engineers, because even easy systems became extremely complex. Besides not all task classes can be reduced to the exchange of physical effects.
Let use take as a little example the incandescent lamp : (simplified and without parameters)
electrical voltage -> [conductor] -> electrical stream
electrical stream -> [filament with resistor] -> increase of temperature
increase of temperature -> [black solid] -> heat radiation
heat radiation -> [ballon] -> absorption
contradictions are not methodical instruments for him
What does creativity actually mean? In my opinion creativity can be reduced to the following basic forms :
Example task : For a long time it is beeing worked unsuccessfully on the development
of "television screens" for blind woman and blind men. Theese "television screens" should enable the blind woman
and blind man and the weak - sighted people to work with computers.
The present technical level is the following : Little steel pins are inserted into cylindrical coils and are
moved upwards and downwards because of switching on and off the electrical stream. Only one single line with
32 signs is made because of the costs. As the speech output is not suitable for complex objects, like for
example drawings, the following tasks have to be solved :
a ) It is to develop a terminal for the presentation of "braille - signs" (braille) with the disintegration of a "Hercules - graphic card" in text mode. The output rate has to be less than 5 minutes.
Even more than 20 years ago it had been possible to exchange the electromagnetic - mechanical process for a better one. There is also no denying, that this solution would bring a considerable effect.
Meanwhile this solution is outdated technically because of a bit thinking over and some experiments, for that reason the next task :
b ) It is to develop a terminal for the presentation of "braille - signs" with the following parameters : It should be able to present graphics with the disintegration of a workstation (1152 x 900). Furthermore it is necessary to present something like a gray scale. The refresh frequency should be about 10 Herz. Besides this solution has to be completly compatible with the LCD - displays (SPARC - book).
Whereas mechanical motions in the classic sense are used in the first solution, we have to take into our heads something better for the second task. In case of the last solution the present known element, the tactile sensor, is maintained, too. I mean with the known way of combination the mechanical motion of the tactile sensor. Actually this task could be classed with the group combination of known and unknown elements in an unknown way. In this case the unknown way is the manner of moving the tactile sensor.
Example task : Imagine, you spend adventure holidays in the Amazon Basin. One week before the end of your holiday you discover a river, which was unknown for you up to now. Of course, you are very bent on making a map of it and on determing the depth of the river. The jungle river is approximately 100 kilometres long. Your helicopter has still fuel for 300 kilometres. The next airport is 400 kilometres away. That is, you are on your own. You have enough film - and photo sets but no measuring intruments for the determination of the depth of the river. In the river you can find sandbanks and trees, which are fallen down into the river. You are able to look over the river from your helicopter and brown water flows in it l ike in the Rio Negro. Without doubt, the trigonometry is the "known way". Only unknown is, on which elements should we use it.
made with SuSE LINUX 9.1, C++, QT, KDE, Apache, Perl, PHP, MySQL and Quanta Plus