Re: improving the UX with the default KDE installation
> Many of the maintainers, including myself, also read this list, so if you
> prepare a diff of the changes with a rationale for each change we'll
> consider them. But please take into account that statements like "I don't
> use it" are no reason enough to drop a dependency as the specific tool
> surely provides a functionality to the desktop user.
As to what I did say myself, as I said at the end of the post, it was mostly
personal feeling, and not rocket science at all. I thought my POV could be of
some use; nevertheless it remains mere POV. Only reluctantly I posted it for
that reason. I've tried to put disclaimers (I might have used bigger font).
This discussion necessarily attract poor quality comments (like mine);
nevertheless the discussion is not completely unworthy.
However, whenever there is a "depend" dependency, it's difficult to make one's
way around.
Also, I may not be very good at doing my installs, and so the consequences are
all my fault: but it often takes me days to remove tons of installed packages
I have no use for; plus finding out how to do it without breaking the system.
Determining something lean enough to be unobtrusive or manageable for further
user change/configuration. But integrated enough to be working seamlessly
right out of the box.
Taking the example of kcalc. It is very small and so completely unobtrusive.
But supposing in the future its development slows down, and it depends on
outdated libraries. With a dependency to "depend" I will have to do with it,
anyway. And it's not as if the puzzle wouldn't make sense without it. So I
suggested it could have a dependency lower than "depend".
The dependency network of a large system like Debian is no simple affair.
Improving it is even trickier. I'm very happy with Debian as it is, and I
couldn't dream of significantly improving it all by myself.
(263 words of idle talk)
>
> Happy hacking,
Reply to: