https://bugs.kde.org/show_bug.cgi?id=399308
--- Comment #17 from Michał Dybczak <michal.dybc...@gmail.com> --- (In reply to Michail Vourlakos from comment #14) > > In my understanding, a pragmatic way to approach this is different. Plasma > in most of its parts is following UNIX philosophy, "do one thing and do it > well". Latte of course does not fall in that category and so does AWC, and > that is ok with me actually, plasma code quality and semantics must be at > the highest degree always. > > An example of this is: https://phabricator.kde.org/D15814 > > even though it has a green light, I dont commit it because I dont feel from > the discussion that it has the degree of quality that it needs to. > > > So what can be done? > I propose different plasmoids: > 1. A plasma plasmoid for global menu (that already exists, and we just need > to maintain it) > 2. A plasma plasmoid for active window title that could provide some basic > options for its style and what it would show when there isnt an active window > 3. A plasma plasmoid for active window titlebar buttons. This is absolutely > possible as demonstrated at: https://bugs.kde.org/show_bug.cgi?id=399757 > 4. A plasma plasmoid separator e.g. a line. That could have some options in > its criteria when to show. I don't think that Unix philosophy is applicable to modern day desktop systems. What we associate with Linux is freedom and lots of choices (also fragmentation) and that destroyed completely Unix motto. Desktops are complex things. In a way, compartmentalizing things is important for order and good management so each part may be focusing on doing its thing the best, but a contemporary, average project is past way this "one thing done right" anyway. Also, from a user perspective, I prefer to have all things in one place instead of being divided and separated. This is also a difference between mature and big commercial projects and many small Linux projects. Linux often feels inconsistent because each project is separate and we build a bigger piece out of it. So here and there, things are out of place and lack of consequences. If a huge project is more complex, bigger and managed by one team, it's easier to have more coherent behavior, look and thus maturity and intuitiveness. Of course, those bigger projects must be founded and managed. Small, free community projects are staying divided and bit chaotic. I like to have one widget/tool to manage those panel and window elements instead of having various parts. For one, global menu from AWC is cooperating with other parts (buttons and titles) so it can dynamically change place. If I turn off AWC's global menu and use system global menu, it stays static in one place and the whole is just dead, unresponsive, limited. This is also the reason that I endure older and bugged implementation of global menus in AWC instead of using better working system one. So the question is if we had those various widgets, could we make them behave as one unity with options that are governing a behavior across the parts that are not part of that single widget? I don't think it would be intuitive to have in one widget an option to set behavior against another one. If we maintain a single widget that does it all, like AWC, this is very possible. In general, AWC is awesome. It needs only an update and a bit cleanup and UI improvements. Finding out working settings is a pain because many combinations don't make sense and produce a gibberish result. That's the main issue with AWC and that should be improved. However, I prefer to have this hardship and still have options, instead of dumbing it down to easier manageable but limited result. This is a KDE/Plasma long-term user speaking here. Although simplifying is a good thing, we still prefer to have options, even if it would create some confusions. With time, confusing elements can be made more intuitive, but having no option leaves us with a dead place. We already have it in Gnome and we don't want Plasma to go that way. The reason Plasma is the way it is, because of options. It is its greatest strength but also a weakness and that is what we love about Plasma. So AWC is great and although it can be improved, I would prefer to have a complex widget with all in one settings. It doesn't matter if we change a name after forking it, but forking it and creating a mess of many confusing, new widgets don't seem to be a helpful or nice idea. We already have enough fragmentation so we love to have bigger projects that stay consistent. Compartmentalization is good if it's done on an organizational level, not on a user level! This by the way why we have DEs. Having one bigger project is better from a usability point of view. If Unix motto was useful in many cases, we all would stay on Openbox, i3 like setups and Windows wouldn't be a thing. -- You are receiving this mail because: You are watching all bug changes.
