encountered_problems.txt

1) understanding that we make the approximation that the growth rate is the same for the whole population at a given size
2) defining priorities: man-machine interface ? models ?
3) accepting the fact that I can't understand the whole project
4) dealing with the number of parameters (vector ? attributes ? both ?) : for calculation functions accessing the attribute directly is better : ndon't need to search in the vector each time. But to set a value to a parameter, a vector is better

C++ problem : pointer toward an abstract class instance in a method gives an error : not declared in this scope
              pointer toward an abstract class instance in the main : no error
WHYYYYYYYYYYYYYYYYY ?

n-2) infinite condition matrix
n-1) dealing with new techniques like matlab engines
n)re initialize all data if two commands are typed in the same main loop
6) tridiagonal linear system (S3), why Gauss-Thomas algorithm more than another one : cf wikipedia
+ strictly dominant matrix in our case