La logique informatique
(selon les travaux de J.D WARNIER)
L.C.S organizing data in a computer system
Study of Operational Logical Files
There will be appearance of Logical Operational Files only and only if:
there is non-empty intersection of data between two or more files.
The data redundancy study will focus on the records of two or more files, and on the data fields themselves.
I remind you first of all, that if there is a non-empty intersection of the records and disjunction of the data fields, or disjunction of the records and non-empty intersection of the data fields, the files are considered disjoint.
I remind you further that the code "identifier" and "Criteria of application" are not taken into account in the intersection study of the data fields.
Here is the schematic of this double intersection: (For example F for the file of external suppliers and C for the file of external clients)
4
We can see that there may be 7 non-empty subsets, which we will call basic logical subfiles.
What can it bring us ?
Imagine that one could identify each of these Sub-Sets, then it would become possible,
to get any meeting of Sub-Sets.
Some may wonder if we are not fucking the flies ! (it's a french expression) , Perhaps, but how to disregard a reality!
This chapter seems to me so important from a theoretical point of view, that I feel obliged to make the following remark:
The problem as it is posed, ie "to make the study of intersection of the FLP" is from a theoretical point of view, absolutely essential if one wishes to control the redundancies existing in our computer system, unfortunately this is a lot of work when you do it manually. I do not think anyone can do that, and we regularly skip this step.
The ideal solution would be : That it is a DBMS that does the work for us, and to do this, we must firstly be able to perfectly identify the real world objects on which the FLP are, and secondly have a dictionary or catalog that perfectly identifies each data field (part of the image on an object), With this one condition this study can be conducted, and above all will allow the DBMS new generation to give us dynamically all F.L.O desired !
An Operational Logical File is a set of data obtained by performing the logical meeting of subfiles.
This model must show :
1°) that all F.L.P must correspond to a "real-world object" and only one example:
- Physical persons
- Products manufactured by the company
- Business
- Company Sections
- Etc ....
The objective is to determine which F.L.P contains images on identical real-world objects, and therefore Suceptible to be in intersection. It will still be necessary to have a common identifier for the F.L.Ps concerned, but that is another story !.
2°) The dictionary of the data fields put in relation with the set of F.L.Ps will reveal the intersection of data fields between F.L.P. If a data field is present in several F.L.P there is intersection (except for the Identifiers and Application Criteria of course).
Come on, I do not mind you any more with the F.L.Os, I just hope has put to you a question on this subject.
But imagine for a moment a dynamically capable DBMS,
to make available to us any Operational Logical File !
And if you are still with me, I propose a nice little synthesis.