Vincent Snijders wrote:
Mehmet Erol Sanliturk schreef:


Unfortunately , there is a concept < law of conservation of bugs >
mentioned in the page

http://www.site.uottawa.ca:4321/oose/index.html#lawofconservationofbugs

Does that mean, we should stop fixing bugs, because you only get more?

Vincent

--



No , it says that fixing bugs will generate new bugs .
Also , please , consider the law about relation between size of software and number of bugs .

Such concepts make fixing bugs a natural situation and this is not a fault of the developers : It is caused by complexity of software and capability level of human beings . The goal is to reduce number of bugs as much as possible in a level that they do not make unbearable damages .


Personally , I am encountering this situation sometimes .


Assume in a program there a three parts :


.
.
.
Part One
.
.
.
Part Two
.
.
.
Part Three
.
.
.


I am getting a bug in Part Two . I do not know that the cause is in Part One . I am fixing Part Two . Part Three is using Part Two . Fixing Part Two is generating a bug in Part Three . I am Fixing the Part Three .


Everything is going well .

Later on I realize that there is a bug in Part One . I am fixing it .
This is causing a bug in Part Two .

I am fixing Part Two .

This is causing a bug in Part Three .
I am fixing Part Three .


This is continuing in chain like the above steps .

The pages listed in my previous message is to emphasize this point :

In a complex software ( which its total comprehension is beyond human capability ) occurrence of bugs is natural and fixing those bugs is an unavoidable step . Then the main point is to eliminate damaging bugs as much as possible .

Another fact is that no one of any complex software is able to stay fixed over time because all of them are used in a dynamic environment .
Changes in this environment is generating a bunch of new bugs to fix .


Then life is going in that way .


If we stop fixing bugs , we encounter < software rot > situation .


Thank you very much .


Mehmet Erol Sanliturk




























--
_______________________________________________
Lazarus mailing list
Lazarus@lists.lazarus.freepascal.org
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus

Reply via email to