1

The KDE interface is good. It has made lots of progress regarding usability. But there isn't a central idea. There isn't a concept unifying everything. And that's the reason why KDE developers feel compelled to make their interfaces as they feel like, instead of being guided by one principle. That shows, and introduces lack of coherence between the interface.

We can use the wise principle of causality to remedy this. Causality, as the relationship between a cause and an effect, can be applied to every KDE program. How? Think about this.

1. We all read from left to right, or from right to left.
2. When I want to do something, I think first in the tool, and later in the result. The usage of a tool (an "Italic" button, for instance) produces a result (italicized text). My tool is the cause. The text, the effect.
3. The causes should be as close to the result as possible. The principle is: I must have a limited, coherent and pertinent subset of functions appliable to only one thing, the thing I'm thinking in, and the thing I want to edit or change.
4. If possible, the logical sequence must follow the natural flow of reading (left to right, or right to left).

Please, review something like this and consider it for your HIG.

The problem with your statement is that it really sounds more like an artistic philosophy than an actual HIG standard.

An HIG is more than what you typed; From the allowable colours in icons, to placing "dangerous" buttons farther away from "safe" buttons, to lining up elements of a form. Cause/effect/reading really has nothing to do with all of these without using several levels of abstraction.

Drafting such a document is worthwhile, and KDE has an HIG already; Google KDE HIG and the first result is the guidelines page.

Operating systems aren't beasts with central ideas necessarily. If you name a central idea, KDE's central idea from the beginning was "a desktop environment, in which users could expect things to look, feel, and work consistently, one my girlfriend can use" (that was the original philosophy, and before KDE, Matthias Ettrichs girlfriend could not use a UNIX computer). The trend for KDE now, the central idea is probably more inline with "draw the power of the open internet"; Plasma is an excellent example, with every almost plasmoid and many applications connecting directly to the web for additional content, information, and networking. Or control panels downloading themes directly from their shells, and all applications updating over the web.

KDE is very usable, and I've seen children under the age of 5 pick it up and go flying. The problem is that people are so used to scraping tools like hacksaws back-and-fourth, while KDE is more like a circular saw - they don't realise you need to push the button to make it go.

This isn't a valid idea, because there is no way to determine when it is done. You should talk to the usability people directly about this.