International System Automation2 / Blog: Recent posts

With 64-length password, it is possible to hide the frequency of appearance of characters within base64 encoded data sequences.

When someone begin to create original own encrypting program on personal computer, what kind of interrupting can be considered?

1)By security update of operating system, replacement of standard runtime libraries
2)By security update of operating system, taking a snapshot of user's filesystem
3)By remote controlling of router, triggering update of operatiing system
4)By taking a movie of Desktop of PC, observation of history of operation

There is a function of a,b and c.

c=f(a,b)

If the f() is "exclusive or", can you get any periodicity in the series of c? Do you know any formula to solve it?

The output series of logic operation of binary numbers can be seen as random series of decimal numbers. But there may be a invisible periodicity within it.

Is it possible to get a and b from the series of c?

$ ./xor
(c:0)=(a:1)^(b:1)
(c:3)=(a:1)^(b:2)
(c:2)=(a:1)^(b:3)
(c:5)=(a:1)^(b:4)
(c:4)=(a:1)^(b:5)
(c:7)=(a:1)^(b:6)
(c:6)=(a:1)^(b:7)
(c:9)=(a:1)^(b:8)
(c:8)=(a:1)^(b:9)
(c:11)=(a:1)^(b:10)... read more

The realy useful cryptographic algorithm is homogenization of byte pattern of plain text, I think.

Is there any good algorithm?

There is no need to be fast. When I have a time to write texts, I have a time to encrypt it. Complete and ultimate are required. Theoretical proof that it can't be read without passwords is really necessary one.

Is there any good algorithm about "concealment of frequency of appearance within data sequences"?
Is there any good method about "homogenization of byte pattern"?

I think, ultimate cryptography is homogenization of data sequences of encrypted data.

The creativity of human being can be expressed as an algorithm.

When you watch a movie, can you feel that you can act as the actor?
Can you play as Toshiro Mifune?
When you see cartoon, can you feel that you can write such things with same way?
Can you draw cartoon like as Katsuhiro Ohtomo?
When you watch Disney characters, can you feel that you can design such things with same way?

If not, your culture-skill is not enough for designing the algorithm of creativity. The algorithm of creativity is easy. You can write a program of creativity, but you can't feel confidence of the algorithm until someone will show a demonstration of the algorithm to you.

int i;
int j;
int k;
int l;
int flag;
int seqlen;
intersection insec;

intersection_clear(&insec);
if( a->num >= b->num ) { 
    insec.num = a->num;
} else
if( a->num <  b->num ) { 
    insec.num = b->num;
}
k = 0;
for(i=0;i<insec.num;i++) {
    for(j=0;j<insec.num;j++) {
        if(a->elem[i] == b->elem[j] ) {
             insec.elem[k] = a->elem[i];
             insec.adr_a[k] = i;
             insec.adr_b[k] = j;
             if(i==j) {
                 insec.distance[k] = 0;
             } else
             if(i>j) {
                 insec.distance[k] = i-j;
             } else
             if(i<j) {
                 insec.distance[k] = j-i;
             }
             b->elem[j] = -1;
             k++;
             insec.the_number_of_same_elements++;
             break;
        }
    }
}
insec.the_number_of_different_elements 
    = insec.num - insec.the_number_of_same_elements;
seqlen = 0;
flag = -1;
for(i=insec.num;i>0;i--) {
    k = -1;
    l = -1;
    k = insec.adr_a[i] - insec.adr_a[i-1]; 
    l = insec.adr_b[i] - insec.adr_b[i-1]; 
    if(k==1&&l==1) {
        insec.the_sum_of_the_length_of_same_sequences++;
        seqlen++;
        flag = 1;
    } else {
        if(flag==1) {
            insec.the_sum_of_the_length_of_same_sequences++;
            insec.the_number_of_same_sequences++;
            seqlen++;
            insec.seq[i] = seqlen;
            seqlen = 0; 
        }
        flag = -1;
    }
}
if(flag==1) {
    insec.the_sum_of_the_length_of_same_sequences++;
    insec.the_number_of_same_sequences++;
    seqlen++;
    insec.seq[0] = seqlen;
}
insec.the_degree_of_similarity =
    insec.the_number_of_same_elements -
    insec.the_number_of_different_elements +
    insec.the_number_of_same_sequences +
    insec.the_sum_of_the_length_of_same_sequences;

The number of combinations to extract 2 pieces from 500 pieces is 125250.
The number of combinations to extract 2 pieces from 1000 pieces is 500500.
The number of combinations to extract 2 pieces from 2000 pieces is 2001000.
The number of combinations to extract 2 pieces from 4000 pieces is 8002000.
...

void combi2(unsigned long long int n) {
unsigned long long int i;
unsigned long long int t;
n--; / n = n - 1 /
t = 0;
for(i=n;i>=1;i--) {
t += i;
}
printf("c(%llu,2)=%llu\r\n",n+1,t);
}... read more

If you are more beautiful and stronger and wiser than me congenitally, you can freely handle this source code and library for any purpose, except for destroying the hierarchy of gene. If not, patience.

When I wrote a mpi/multiple precision integer library by myself alone, I had to propose a such license.

When you want to compare the two set, you have to calculate the value of similarity of them. The value of the similarity can be calculated with below formula. I found out the formula by myself.

the value of similarity = { the number of same elements - the number of different elements + the number of same sequences + the sum of the length of the same sequences }

By this formula, you can make brain-like behavior/action analysis engine. Yes I can

A says "X1 = {a1,a2,a3}".
B says "X1 = {a2,a3,a6,a7}".
C says "X1 = {a1,a2,a4}".

then,

If A=B=C
X1={a1,a2,a3,a4,a6,a7} OR
X1={a1,a2} AND

If A>B=C,A>B>C,A>C>B
X1={a1,a2,a3}

If B>A=C,B>A>C,B>C>A
X1={a2,a3,a6,a7}

If C>A=B,C>A>B,C>B>A
X1={a1,a2,a4}

If A=B>C
X1={a1,a2,a3,a6,a7} OR
X1={a2,a3} AND

If B=C>A
X1={a1,a2,a3,a4,a6,a7} OR
X1={a2} AND

If A=C>B
X1={a1,a2,a3,a4} OR
X1={a1,a2} AND

then, someone may resay something to modify previous classification.

The number of point of view of time-series data consisting of n columns is n * (n-1).

Yes, it is possible to implement automatic table generation per a new classification occurrence.

But elimination of data table have to be automatically done per occurrence of annihilation of classes.

Is there reference implementation of "classification engine"?

There are two types of "classification", one is generation of information, another is elimination of information.

According to human nature, classification engine can't be rational. But it can be explained by rational theorem.

It is impossible to use traditional RDBMS in disputed area where disease and terrorists gathers. Because there is not enough time for IT consultants to do ER diagrams modeling. The automatic data base construction by message exchange is required for that purpose. It is useful for any field science or the place where many kinds of races gather.

In traditional RDBMS, the start-point of classification have to be set for using it. In disputed area where many races gathers, the start-point of classification can't be set.

In "system improvement automation system", if motivation of improvement of organization is high, the quantity of variable message will increase. If not, fixed message will increase.

When a classification is occur, the number of class is always countable.

The number of category may be variable, but it is always countable. It can't be infinite.

There are two fundamental desire about information processing. One is desire that I want to sort the information, and another is desire that I want to grasp all the information.

Human being can't satisfy them simultaneously. They can't focus on both of them simultaneously. They have to be handled step by step in brain.

So, in what situation does such switching occur in your brain? The condition of such switching could be a key elements of next of RDBMS.

The logging record of UNIX Operating System is like below.

{ timestamp, identification of person, selected behavior }

The information format is very similar to historical records by News Papers. The number of points of view will be automatically determined by the number of columns and classification.

The number of points of view can be automatically determined by the number of classification. The number of classification depends upon the velocity of generation and annihilation of classification.

RDBMS may be suited for National Library of Roman Empire, but it is not suited for disputed area where disease and terrorists gathers. Because start-point of classification have to be defined.

A solution for automatic database generation is useful for such area. Indeed, by exchange of such message, automatic database generation is possible.
X={a1,a2,a3,...an}