Well said, Andrew. This is only one option out of many, and I share the view that more could be designed to ship. That kind of choice is why I'm here on KDE in the first place. I look forward to seeing more design ideas here in this thread, and check it daily in hopes that there will be more. We have a lot of great artists and ideas in this community.

alake wrote:At Thomas' (colomar's) urging, I wanted to chime in to share a little bit of the rationale for the design choices for the Breeze window decoration design.

I confess that I'm doing this somewhat hesitantly because this kind of stuff can often be interpreted as being defensive and can often devolve into fruitless argument. But. I entirely agree with Thomas on usefulness of documenting rationale. So here I am. Below is a slightly edited version of a conversation I had with Thomas.

Design is a tricky thing, everyone has their own opinion, I agree entirely and it shouldn't devolve in to a flame war of course. I hope it's not taken that way at least.

----

alake wrote:What is the reasoning for using a color for the windeco which contrasts the main window color? I personally like uniform colors (like in the one from garthecho), so I'm interested in the advantages of a high contrast between title bar and window from a design perspective.
I was sticking with a single color because that's how I thought a window "should" look as well. The downside is that it severely limited the design options for anchoring the window title in the visual hierarchy. It was actually one of garthecho's designs that broke the log jam.

I think it can be done well, if the current window style design were to evolve. I think Chrome OS does window titles pretty well, I've included an example, it draws a clear distinction to the title bar and it seems to be done well.

alake wrote:For a design n00b: What does "anchoring the window title in the visual hierarchy" mean?
The visual hierarchy is just the order in which information in the overall design (in this case the window and it's contents) is intended to be consumed by the viewer. The window title is the starting point in the visual hierarchy for the window, thus the anchor. Which means it needs to have visual qualities necessary for that purpose. While it's not impossible, the common color for the background of both the window and the titlebar, make it somewhat more challenging. That's partly why some of the earlier designs has the title in the highlight color. The problem, as many folks here pointed out, is the difficulty of reading that highlight blue on the near-white background.

I think it's easier to pull off the OS X style titlebar with proper gradients, KDE appears to be moving to a more "flat" design. I think flat can be done well but yes I agree it would be difficult to pull off a unified titlebar + window with a flat design, at least I've not seen any that pull it off well. Again though I do think we could draw some inspiration from Chrome OS, I think they've pulled it off well. I think the trick here is the titlebar color may differ from the window color but it should be subtle. The blue titlebar underline does appear to be a bit "loud" to me, perhaps it is just the color itself though, I've seen this used in Chrome OS and it looks acceptable.

alake wrote:So the question then is: Why do other OSes/DEs seem to make title bars rather subtle? Transparent in Windows 7, blending in with the toolbar in GNOME 3 and OSX...
Good question. One which I pondered as well. What's happened over time with all three is that when the same background color is used everywhere, the titlebar as a separate visual entity from the window itself doesn't seem to make a great deal of sense - acres of space with little obvious distinction in the visual heirarchy. It's all in hindsight but the result is much the same: they're all eliminating the visual and functional distinction between the titlebar and the window contents. It isn't a coincidence that it visually looks better as well.

I'm conflicted here, I think it can be done very well like OS X or done poorly like Gnome 3 (mostly due to the fact OS X has a global menu and Gnome 3 hides menus all over the place..). Visually though I think it looks good, obviously OS X design and widgets trumps Gnome 3's by a wide margin.

alake wrote:Ah, so the color is a symbol of our commitment to the title bar as a distinct entity?
In our case, we're not going to get rid of the titlebar anytime soon (for many good reasons). So rather than try to hide it, after seeing garthecho's Sol design on deviantart, it made sense to me that we own up to the titlebar as a distinct enitity and design for it rather than hide it. When I shared the the first cut at the new design with garthecho via email, he essentially agreed. His thought was that window decorations on all platforms had overreacted to the OS X unified titlebar+toolbar look (I hope garthecho doesn't mind me sharing this publicly). So yeah, like you said, it's a commitment to the titlebar in the visual design rather than trying to hide it.

I think this design could be done right. I don't think it's quite there but I am sure it will improve over time. I think the titlebar can stand out but it just needs to blend better with the windows.. of course that could be handled by a proper color scheme.

As far as the overreaction to OS X, I can understand the desire to emulate. The unified titlebar+toolbar applications in OS X do look pretty slick. Is it imitated well though? Not always. Elementary OS seems to be the closest i've seen so far.

Does KDE have to follow suite? Not really, no. I think KDE can look good with regular ol' titlebars, I am not advocating a Gnome 3 style windows.

----

alake wrote:Now let me share a few other thoughts. This does not mean that I think that window decoration designs wherein the titlebar shares the window background color are somehow wrong or are incapable of being beautiful. This does not mean that I think this window decoration cannot be improved upon (in fact, that's already happening). I firmly reject the notion of One True design. I firmly reject the notion that design only moves forward on One True path. Those notions are anathema to the inherent creativity of any design effort, visual or otherwise. More importantly, I think this community is capable of coming up with fantastic window decoration designs that are in no way obligated to satisfy the rationale shared here for this specific design.

I think it can be done right. Here's the bad thing about not having a "One True Path" though, or at least a very attractive default theme.. KDE is often maligned for being "old and ugly". A great community is fantastic and I don't dispute this but KDE 5 really needs to have a great default theme to lure back users. I cannot count how many times I've preached the great aspects of KDE and in return I'm told "yeah that's great but KDE is ugly".

alake wrote:Remember this is for Plasma. The KDE desktop environment. KDE. We, as a community, have always loved options. It is well within our wheelhouse to ship multiple decorations. Always feel free to create great designs!

I love the ability to customize, I don't think those options should ever go away. Honestly I still like Oxygen, I will probably continue to use Oxygen myself unless the new themes blow me away.

For me the problem isn't that there is a title bar. It's that the way that it's designed it draws too much attention to it.
Owning up the titlebar as a distinct entity and design for it shouldn't equate that it is design in a way that unnecessarily draws attention from the users of the content on the window.
An improvement of this design would be to strike a better balance between having something that shows that thoughtful design was applied to it but that it isn't screaming "look at me" and that it blends well with the rest of the Plasma desktop components.
The title bar for me draws too much attention to it because of the conjugation of the blue line, the dark color that contrasts with the rest of the desktop and the distinction between the close button and the other buttons on the bar which is completely pointless from a functionality perspective.

As for the line of reasoning that more options can be included, for me that can't serve as an excuse because what truly matters for 90% of the users is the default look. The default look is what matters and not how you can customize because most users don't much more customization than changing the wallpaper.

prosmaninho wrote:distinction between the close button and the other buttons on the bar which is completely pointless from a functionality perspective.

The close button is different from the other buttons:
- It's presumably the most used one
- Closing the window often closes the application as well, which does make it different from just minimizing or maximizing the window

prosmaninho wrote:For me the problem isn't that there is a title bar. It's that the way that it's designed it draws too much attention to it.
Owning up the titlebar as a distinct entity and design for it shouldn't equate that it is design in a way that unnecessarily draws attention from the users of the content on the window.
An improvement of this design would be to strike a better balance between having something that shows that thoughtful design was applied to it but that it isn't screaming "look at me" and that it blends well with the rest of the Plasma desktop components.
The blue line for me draws the eyes too much to it (unnecessarily), the distinction between the close button and the other buttons on the bar is also completely pointless from a functionality perspective.

As for the line of reasoning that more options can be included, for me that can't serve as an excuse because what truly matters for 90% of the users is the default look. The default look is what matters and not how you can customize because most users don't much more customization than changing the wallpaper.

Well said, can't agree more.

The average user looks at the window manager and window content as ONE thing that belongs at a functional level together, he doesn't distinguish between the two and therefore there shoudn't be a visual distinction.

What's important for him, is the content of the window, the application itself ... the toolbar and window elements like close- and minimize buttons are accessories, neccessary, but still accessories. If anyway, then these two elements should be visually grouped together, because they are 'belonging' to the main content of the window, and as they do, they are dinguished from it.

Isn't the purpose of visual design (beside beeing beautiful and light on the eyes) to represent the functionallity and lead the user to the parts of the application, that are important to him? I guess, the windowbar is the least important part of the window for him, therefor imho it shouldn't be represented as it would be the most important part.

I agree with Uri that the title bar should have something to distinguish it from the rest of the border but not by using so many elements to cause that distinction that draws the attention of the user to it instead of the content. IMO, use only one design element to indicate one thing. To distinguish the window title from the rest of the window use just the blue line as separator or the black background, in the minimize/maximize/close button use just button shape, etc. Too many elements cause confusion, draw attention of the user from the content in the window to something that is supposed to be out of the way unless if people are looking at it.

Furthermore, there's one thing that annoys me in the Gnome DE, KDE 4.X, Cinnamon, etc : that the window style is always different of the taskbar in the DE. It gives a disjointed look to the DE and one thing that I think that Canonical actually did well in Ubuntu with the Ambiance theme is that it doesn't do this mistake. Plasma desktop has a chance to avoid that mistake as well.

Therefore, I tried to make a design for the window style that combines well with the taskbar style of the Plasma desktop so that the window style matches the look of the other component that people use for window management - the taskbar:

Window Style:


Taskbar


This is a rough mock up, mind you so please be gentle.
Just to better pass an idea of what I meant with using less elements to distinguish the title bar from the rest of the window and providing a tighter integration with the rest of the Plasma Desktop UI.
1 - Keeping only the blue/grey line as a separator of the window title with the rest of the window,
2 - Plasma blue color for the separator on the active window and greyed out with the program title in inactive windows (maybe also the minimize/maximize/close buttons?),
3 - Color of the window title equal to the color of the window.
4 - The blue/grey separator shouldn't cut the window title from the rest of the window, just use it as a separator for the window title from the rest of the window but integrating it with the window instead of being something on top of the window.

I used the window buttons by mmistretta in this mock-up because I liked his button design and didn't have time to try out others. Trying different fonts would also be cool as well as justifying the text instead of centering it.
I am sure that any of you could come up with a better design this, so I would love to see something that stuck with Herrera's principle of making the title bar more proeminent but with only one design element used to achieve this and to describe the button actions. You can make the window bar elaborate but with few design elements not making it look cluttered.

The close button is different from the other buttons:
- It's presumably the most used one
- Closing the window often closes the application as well, which does make it different from just minimizing or maximizing the window

Colomar, that's just grasping at straws. It is a button that performs a different function than the others so it has a different shape already - a cross.

I don't believe you should distinguish it from the others because it is the "presumably" the most used one (is it? - I know that in my use case I use the minimize button much more often than the close button since I tend to jump around between a lot of open, active windows) - is that reason for distinction used in other use-cases or you want to force your presumption on others?
Going down that slippery slope leads to the way Gnome does things.
Everybody knows what a close button does - it doesn't justify being highlighted from the other buttons. If that's a reason for distinction then make it a general rule to all the buttons and do it like MacOSX does with the green, yellow, red color for the minimize/maximize/close button. Again it is just more visual clutter to describe what the button does when you already have the x shape to define what it does. Good design gets out of the way - it doesn't scream "look at me".

I just want to end my post by stating that I love the work you guys are doing here. I love Uri's work on Nitrux, and the work he's been doing here so that's why I hope that someone with his talent will listen to my small input and further improve the Breeze window style or come up with an alternative that serves as the default window style.
The System settings work is also tremendously impressive untill now, as is the new mouse icon sets, etc, etc.
This is my small contribution to you guys. I don't want to ruffle feathers but provide constructive criticism (hence why I pointed specific issues and provided this sketch as an indicator of what I meant).

prosmaninho wrote:Furthermore, there's one thing that annoys me in the Gnome DE, KDE 4.X, Cinnamon, etc : that the window style is always different of the taskbar in the DE. It gives a disjointed look to the DE and one thing that I think that Canonical actually did well in Ubuntu with the Ambiance theme is that it doesn't do this mistake. Plasma desktop has a chance to avoid that mistake as well.

Humble suggestion.. I know this isn't a Plasma thread but this would be my idea for Plasma Next theme. I would go with an icon-only transparent taskbar. Unfortunately with the current plasma the Icon Only taskbar does not apply any drop shadow effect to the icons/system tray, I the icons would look better if it did.

Example (my desktop):
http://postimg.org/image/nhs9al9bd/full/

Ok just a tiny tiny thing here Prosmaninho:
this work has been ongoing for months. We have never been scared of user input and have been very forthcoming in letting ANYONE join in the work (the majority of the people doing the work are people who just joined in). The current window decoration is the... what tenth version? I don't think I am being overly dramatic by saying that - it has changed and always because of community input, new awesome ideas etc etc. There are too many contributors to count who have laid if not their hand but their input on it.

This last version is the result of this entire thread but also email exchanges, debates, blog-debates and chats. Its passed usability experts, designers and developers. I can say with the ten page thread we're talking in as backdrop - this is the one we're going for.

I am not saying your suggestions are bad - or that they are not welcome (they are and dude, you are awesome for posting them, don't you dare think otherwise) but I can say with objective certainty that "or you want to force your presumption on others? Going down that slippery slope leads to the way Gnome does things." is uncalled for. Not just because I think using Gnome as a pejorative is fairly low to the Gnome designers but also because Thomas have been nothing if not flexible every step of the way. For a guy who could have, at any point just said "Pfft read the HIG" or referred back to logs of previous choices and decitions he has instead been forthcoming, open, kind and made certain to never "pull rank" or for that matter documentation on people.
Same with everyone else in this thread - it has changed over and over due to the input and with that measured towards usability and design goals and never once has an idea been disregarded out of hand.

(I wont comment on the "slippery slope" argument for now)

And there are still options as Andrew pointed out. The dark title bar is one of two options AND this work wont even ship per default in the Beta (since window decorations demand some C++ work as well and the Aurorae theme has raised an issue there). So there is absolutely no reason to try to, in all fairness, guilt someone else in the community for not changing his mind at the drop of a hat.

My point is this - we need to [edited due to censored words] "Poop or get off the pot" at this point. We can't keep skipping from one foot to the other. And even then, even if this ships, its never set in stone. Work can and should continue indefinetly.

I can only suggest you check back earlier in this thread with the previous ideas - you can find a ton of ideas and work that you can expand on and we can have even more window themes in Plasma to change to if we feel like it.

I hope you don't consider this a snub - its not. Consider this an open invitation to join in - just that we have a deadline and as much as we should always appreciate and respect people like you who spend their free time writing great input we have to make certain that the work already done is also respected and appreciated.

Sorry I haven't been able to read everything so this may be pointless.
Has the line been tried above the title and the same length as the title? (Well, one pixel longer on each end.)



Not wanting to argue but, the line is above the window title in the default position of the task manager.

Such may be too hard to code but, it would be neat if the title and task manager both did this.

jensreuterberg wrote:Ok just a tiny tiny thing here Prosmaninho:
this work has been ongoing for months. We have never been scared of user input and have been very forthcoming in letting ANYONE join in the work (the majority of the people doing the work are people who just joined in). The current window decoration is the... what tenth version? I don't think I am being overly dramatic by saying that - it has changed and always because of community input, new awesome ideas etc etc. There are too many contributors to count who have laid if not their hand but their input on it.

This last version is the result of this entire thread but also email exchanges, debates, blog-debates and chats. Its passed usability experts, designers and developers. I can say with the ten page thread we're talking in as backdrop - this is the one we're going for.

I am not saying your suggestions are bad - or that they are not welcome (they are and dude, you are awesome for posting them, don't you dare think otherwise) but I can say with objective certainty that "or you want to force your presumption on others? Going down that slippery slope leads to the way Gnome does things." is uncalled for. Not just because I think using Gnome as a pejorative is fairly low to the Gnome designers but also because Thomas have been nothing if not flexible every step of the way. For a guy who could have, at any point just said "Pfft read the HIG" or referred back to logs of previous choices and decitions he has instead been forthcoming, open, kind and made certain to never "pull rank" or for that matter documentation on people.
Same with everyone else in this thread - it has changed over and over due to the input and with that measured towards usability and design goals and never once has an idea been disregarded out of hand.

(I wont comment on the "slippery slope" argument for now)

And there are still options as Andrew pointed out. The dark title bar is one of two options AND this work wont even ship per default in the Beta (since window decorations demand some C++ work as well and the Aurorae theme has raised an issue there). So there is absolutely no reason to try to, in all fairness, guilt someone else in the community for not changing his mind at the drop of a hat.

My point is this - we need to [edited due to censored words] "Poop or get off the pot" at this point. We can't keep skipping from one foot to the other. And even then, even if this ships, its never set in stone. Work can and should continue indefinetly.

I can only suggest you check back earlier in this thread with the previous ideas - you can find a ton of ideas and work that you can expand on and we can have even more window themes in Plasma to change to if we feel like it.

I hope you don't consider this a snub - its not. Consider this an open invitation to join in - just that we have a deadline and as much as we should always appreciate and respect people like you who spend their free time writing great input we have to make certain that the work already done is also respected and appreciated.

I understand that and thank you for reading my feedback and replying. I know that this is a lot of work and if a version is more or less settled it would be a pain to go back to the drawing board and have something that would reach a consensus.
I really enjoy the open way in which you are doing this work and wanted to give you guys my contribution in any way and not just pass as a jerk that knocks down your work without even offering constructive feedback. Therefore, I tried to make a rough sketch based on the principles that I was trying to convey since a few other people disliked the design.

And when I mention Gnome, I am not mentioning it as a pejorative to their design work (which is well thought) but their "my way or the highway attitude" that dismissed and alienated user feedback (even leading to multiple forks of the DE) and that fortunately KDE isn't following. It's just that his reply to me seemed to be so dismissive that it felt that way and that is what I was referring to. I wouldn't even have wrote that if his reply wasn't there. I am not mentioning this in reply to the window title work that has been presented here by others and that has been the result of consensus.

Prosmaninho:
Do what I do, chalk it up to cultural differences and assume people are being nice by default

(this may sound silly but in the last half year dealing with people from more countries than I even dare count, it really has helped. Take Germans for example. In my first experiences I thought they where being rude and (to be totally honest) a tad thick. Then I realized that in fact the way they communicated was close enough to my way of communicating but with the slight difference in that they didn't grasp my small communicative ques and I didn't grasp theirs - that I couldn't spot it. It was too close to home. When this project is over (or rather my involvement in it) and done I can say that this has been one of my biggest educations - talking and working closely with people from every continent on the planet, people I often don't even know what their real names are. As for me always assume that everything I write is written smiling - I can be overly dramatic and bombastic but I'm happy being it)

As for work on the Window Decorations - I think that this is one of those points where we have to wrap it up. It IS unfair to those of you who just joined us but the alternative would be too complex at this point. I wish I could figure a clever way out of it - but we need to go "Ok we've pleased as many people as possible, lets go to work on the rest of the DE". Again, I sincerely wish it wasn't like that - but we have to.

The alternative I think is to start a new thread about window decorations titled "Future Window Decorations" to let people hammer at new ideas for it for future releases.

I am asking this in the most humble way possible: what do YOU think about that?

jensreuterberg wrote:Prosmaninho:
Do what I do, chalk it up to cultural differences and assume people are being nice by default

(this may sound silly but in the last half year dealing with people from more countries than I even dare count, it really has helped. Take Germans for example. In my first experiences I thought they where being rude and (to be totally honest) a tad thick. Then I realized that in fact the way they communicated was close enough to my way of communicating but with the slight difference in that they didn't grasp my small communicative ques and I didn't grasp theirs - that I couldn't spot it. It was too close to home. When this project is over (or rather my involvement in it) and done I can say that this has been one of my biggest educations - talking and working closely with people from every continent on the planet, people I often don't even know what their real names are. As for me always assume that everything I write is written smiling - I can be overly dramatic and bombastic but I'm happy being it)

That may be the case, and it's also another thing: we are communicating in written form so it's always hard to understand what emotion the other person is conveying (if he's being nice or sarcastic, etc).

As for work on the Window Decorations - I think that this is one of those points where we have to wrap it up. It IS unfair to those of you who just joined us but the alternative would be too complex at this point. I wish I could figure a clever way out of it - but we need to go "Ok we've pleased as many people as possible, lets go to work on the rest of the DE". Again, I sincerely wish it wasn't like that - but we have to.

The alternative I think is to start a new thread about window decorations titled "Future Window Decorations" to let people hammer at new ideas for it for future releases.

I am asking this in the most humble way possible: what do YOU think about that?

Well, I think that's a nice gesture for sure.
However, the importance of defaults to cause a good first impression when someone is introduced to a unfamiliar piece of software (any type of software) is paramount. The first impact that I got out of that window decoration wasn't as positive as what I've got from all the other elements of the work that you guys have been developing here and in fact it seemed to contradict the rest of the philosophy and that's why I wanted to give my small contribution.

Either way, I think if the people that contribute the most are pleased with the result then they are completely entitled to move forward if they feel like it. Especially, when there's a tight schedule to work on so many other things as well.

prosmaninho wrote:And when I mention Gnome, I am not mentioning it as a pejorative to their design work (which is well thought) but their "my way or the highway attitude" that dismissed and alienated user feedback (even leading to multiple forks of the DE) and that fortunately KDE isn't following. It's just that his reply to me seemed to be so dismissive that it felt that way and that is what I was referring to. I wouldn't even have wrote that if his reply wasn't there. I am not mentioning this in reply to the window title work that has been presented here by others and that has been the result of consensus.

Sorry for not just letting this rest, but I feel the need to clear things up. Since your "slippery slope to Gnome design" remark was in reply to my post, I assume that mine was the one which "seemed to be so dismissive it felt that way". In that case, it indeed must have come across completely wrong. I didn't mean to dismiss your input at all! All I wanted to offer was an explanation of why it could have been done that way, to counter a point you made which I felt rather disrespectful towards the work, and, frankly, very dismissive on your part: "distinction between the close button and the other buttons on the bar which is completely pointless from a functionality perspective.".
When someone says that part of the work of someone I respect (alake in this case) is "completely pointless", I feel the need to make clear it isn't. Yes, the points I made for it were not very strong, but even weak points are enough to show something is not "completely pointless".
Actually, I think you're lucky that alake has that remarkable ability to not feel offended. I've read so many comments to his work in this forum which I'd have considered very disrespectful, but he doesn't seem to mind at all. I can tell you that if I have an idea and you go ahead and tell me "it's completely pointless", I'm going to lose my temper, as this does not show any respect at all.

Again, all I wanted to express was "a distinction between minimize/maximize and close isn't completely pointless", which I still maintain it isn't. Just because something is not completely pointless does not mean that it's the best thing in the world, it shouldn't be changed and thus any input to it is dismissed, though. The constructive part of your post (especially the mockup) are very welcome, this is what this thread is for!
So no, we do not have a "This is our design so stfu!" attitude (otherwise we wouldn't even have the public forum in the first place!), but Jens wrote up the "How to comment, criticize and give feedback!" rules for a reason. Maybe I should just have pointed to those rules instead of trying to defend someone else's work, I just felt it would be more constructive to point out possible reasons for this design choice. Okay, I admit, that didn't go as planned. Sorry.

So please, keep your ideas and constructive feedback coming, just maybe try to avoid calling things "completely pointless" in the future. "From my perspective, the differences between minimize/maximize and close are not strong enough to warrant a different design for the close button" would certainly not have triggered that response from me.

So if I can join in for a moment as well:

Unlike prosmaninho, I like the different look of the close button. I guess I agree with colomar's reasoning for it - or perhaps I'm just brainwashed by the 4 last versions of MS Windows.

However, I'm more in agreement with prosmaninho when it comes to the Title Bar. I thought that the earlier design suggestions with the highlight coloured line on top and no difference between the bar and the rest of the window looked pretty neat.

Now, I've been reading and rereading alake's explanation for it (http://forum.kde.org/viewtopic.php?f=285&t=120084&start=120#p310158), but I'm still not convinced. But Jens says that this more finalised design has "passed usability experts, designers and developers", and I freely admit that I'm not an expert here, but I really don't understand how this design is better than how it seemed like it would turn out earlier.

I guess my final point is that usually when I disagree with a well thought out decision it is because I haven't yet grasped the reasoning behind it. So could anybody please try to explain again why the Title Bar should be so prominent? Or point me to somewhere I can read more about it? Cause in all honesty I don't like the distinct Title Bar at the moment.

My reason for liking the distinct design for the close button is because accidentally pressing it can have the most devastating effect. It's similar to nature showing bright colors to indicate poison.