Le 23.12.2012 03:37, John Hasler a écrit :
berenger.morel writes:I really would like to understand why people think it is a problem todo softwares able to run on lower hardware... if someone have any clue, I really want to know it!Because they aren't very good programmers. And that's a problem,because there is more programming to do than there are good programmersto do it. -- John Hasler
But what I do not understand is why it is starting to become a rule to avoid performances. Sounds like people think it is a good thing to do the job of the OS or the WM in a simple application. Now, we are starting to have HTML interpreters, aka web browsers (just an example), able to manage windows, load and execute programs, access to webcams... and mozilla is building OSes...
I'm not a good programmer (or at least, I do not think to be good enough to claim such thing), but I know that I am not able to build an OS (well, I must admit that is might because I tried in my first years of programming ^^). I do not understand why people are taking what is obviously the wrong way to do things: reimplementing features which are already given by other softwares, in applications which are not specialized in those features. The most obvious example is window management, I think. Most softwares nowadays implement window management features, like tabbed and MDI interfaces (and we can see consequences: memory leaks, crash which makes you loose all your work, strange behaviors, ...) And the post from which I replied was speaking about people who just think they can manage disk access better than the kernel!
Is not it non-sense? Being a good programmer or not is not the problem here, even a newbie could understand that multiplying features in a software makes it harder to maintain and memory hungry... no?