Slashdot Mirror
GNOME To Lose Minimize, Maximize Buttons
An anonymous reader writes "When GNOME 3 arrives in a month, users might be surprised to see old UI staples 'minimize' and 'maximize' buttons gone and replaced by... nothing, in the case of minimizing, and either drag-up or double-click-titlebar for maximizing. Says Allan Day, GNOME Marketing Contractor: 'Without minimize, the GNOME 3 desktop is a more focused UI, and it is a UI that has a consistent high level of quality. Yes, moving to a minimiseless world might take a little getting used to for some, but the change makes sense and has clear benefits.' Some users already welcome the change, while others are in an uproar, swearing to wait for GNOME 3.2, switch to KDE or even Windows. What do you think? A better, simpler interface for new times, or a case of making something simpler than it should be?" I like minimize and maximize buttons, but I'll admit to liking the look of GNOME 3 .
The world becomes more and more like satire every day.
Deleted
Windows: focus groups, study users, never get it quite brilliant but basically give people what they want.
Apple: try to think about what will appeal to the user, deliver to maximise experience.
Gnome: WE DID COMP SCI IN COLLAGE AND HURD OF DON NORMAN THIS MAEKS US EXPURTS ON UI DESIGN. WE HAVE NO EVIDENCE OR TRACK RECORD BUT U WANT WAT WE WANT. WE WANT TO MAKE A NAME 4 OURSELVES PLS ACCEPT ONE OF OUR IDEAS PLS!!!
For ubuntu to drop Gnome for Unity.
Blender And Linux Fan
Running Ubuntu 10.10 with a gnome-shell build from the git repository, and I have to say, I love the change. The minimize and maximize functions themselves are not gone; as the summary says, you can still double click the title bar to maximize. If you want to minimize, you can right-click the titlebar, then click minimize, or using ALT+F9.
I think this is a great design change. In Gnome 2.0 and less, like windows, you would minimize windows to make room/less clutter for windows you're actually using at that top. Now, instead of minimizing, a better method is to move it down a screen (right-click, move to workspace down), or zoom out to activity view, drag the screen from one screen to another. I find when I have a lot going on -- multiple browser windows, terminals, ftp client -- I use this a lot. It comes in handy being able to separate each website you're working on, each server, into it's own workspace, free from intrusions and other unrelated stuff.
Good job Gnome devs!
It seems like both the KDE and the GNOME folks have decided that they need to reinvent this whole desktop thing. KDE decided that icons are unnecessary, now GNOME deems maximize/minimize buttons unneeded.
Guess I'm lucky to use IceWM which still works the way it worked ten years ago - and I find that a good thing.
Real life is overrated.
Tested in the real world? No, not at all. But the developers have done a lot of reading of theoretical papers, so how could this go wrong?
Dragging is more stress-inducing to the hand than simply clicking mouse. we do countless minimize-maximize actions over the course of a normal workday.
I cant risk more potential for RSI, just because a few people think that is better to do so.
Excuse me gnome, but you are losing me.
Read radical news here
On a day to day basis I rarely use any of the titlebar buttons. I double-click the titlebar to maximize/restore, click the taskbar or dock to minimize, usually just punch Alt+F4 or Alt+Q to close.
And the masses cried out, "09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0!"
I like clean interfaces, but seriously, what does this help? It doesn't save space, the title bar is still there. Ignoring those buttons costs nothing, and replacing a button with a non-graphical multiple-action like double clicking isn't making an interface simpler, it's making it more complex. I understand the confusion about a minimize button with no taskbar, but this doesn't seem like a particularly well thought out design change. We got rid of feature X, so action Y isn't the same anymore. Okay, just get rid of it.
"Until the become conscious they will never rebel, and until after they have rebelled they cannot become conscious"
UI tasks enjoyable? What drug is this blogger on?
A UI's purpose isn't to be enjoyable, it's to let the user do what he wants/needs to do and otherwise stay out of the way.
Case in point: Clicking a button is going to be a lot quicker and require me to do less thinking than dragging a window around.
Having said that, I like snap and would like to see several of its features included, but not as the primary replacement for the maximize button.
GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
When are we going to get an interface that is totally configurable to user preferences?
Someday, I'd love to sit down at a computer, point it to the URL where my interface preferences live, and presto - it instantly becomes the desktop I'm most familiar with.
Think of it as the GUI equivalent of setting your shell in .profile.
"Slow down, Cowboy! It has been 3 years, 7 months and 26 days since you last successfully posted a comment."
The hope is that the developers will put more attention to the meat by fixing the numerous bugs that are lurking into the GNOME suite.
Otherwise they'll end up with a new KDE 4.0 fiasco.
Anyway, hiding buttons is not a real great advance in my humble opinion.
Maybe Computers will never be as intelligent as Humans.
For sure they won't ever become so stupid. [VR-1988]
I use Windows 7 and never, ever use that. In fact, it pisses me off that a good 10% of the time when you're just re-arranging windows, it wants to maximize something instead. I'd be interested in knowing why you like the click and drag over clicking a min/max button, which means taking a good 300% more time to maximize a window.
"The tree of liberty must be refreshed from time to time with the blood of patriots and tyrants." ~Thomas Jefferson
I haven't used Gnome 3 so I don't know if I like this change. But I have one request for the devs: ***PLEASE*** make it *easily* possible to retain the Gnome 2 look and feel if a user prefers that. TFA wasn't clear about whether this would be possible.
You become comfortable working in a particular way. Then you upgrade ---all your reflexes are wrong and you have to waste time relearning the interface. If I'm productive, let me stick with what I know. For a developer to alter the UI without a downgrade path (as MS did with the Office ribbon) is the height of solipsistic arrogance.
I love Linux, but it's like everyone mutually agreed to abandon desktop sanity. KDE never met an option they didn't like, and Gnome never met one they did. I've used both extensively and recently but both make me spend more time cussing at the screen than I want to. I've held on to Linux (and FreeBSD) desktops for over a decade but I give up. It's not going to happen. I'm still going to work on a Unix all day, but I'm switching to the pretty one.
Dewey, what part of this looks like authorities should be involved?
This is such a drastic step of changing a UI paradign that's existed for the past 25+ years, and the only justifications I see for it are completely theoretical ones. Where's the usability testing by actual users to see if the theories hold any water?
Both sides can argue about what THEY think the user will prefer. The arguments can sound extrodinarily convincing, but what actually matters is how it performs in the real world with actual users. The solution to this problem seems to be "just put it in the next release and see if people revolt enough" rather than conducting actual controlled tests. IMO this is an extrodinarily flawed approach. A controlled test gives you non-biased opinions rather than political ones. This approach only seems to create a rift between the two opposing sides rather than finding out what's the best UI experience for the user.
AccountKiller
Minimize is removed because the concept doesn't make much sense in GNOME Shell. Minimize only has an intuitive function when used with a panel, while in GNOME Shell all it does is make the window disappear. The last time I tried GNOME Shell, minimizing did prove to be a frustrating habit acquired by years of having a panel.
Maximize on the other hand is removed because... well, because this is GNOME we are talking about...?
They're arguing that minimizing is an uncommon thing to do and also one that doesn't work well within the general interface ideas behind Gnome Shell. So minimizing is, basically, deprecated. OTOH, they're not at all saying that maximizing is infrequent. What they are saying is that you should maximize in other ways: primarily by dragging the window to the top edge (that'd be the same as in Win7; the mouse gesture might be different, I haven't really tried Gnome 3); double clicking the title bar will also still work, I assume. Mouse gestures are supposedly more "gratifying" or some similar thing that will undoubtedly get a lot of hate on Slashdot.
FWIW, it's true that I only really use the close button on the title bar. I rarely minimize windows, and I invariably maximize by double clicking the title bar.
Switch back to Slashdot's D1 system.
... always worked as an alternative for maximize/restore on MS Windows. Just like double clicking the top-left corner closes a windows since Windows 3 at least.
I'm thinking, for example, about Firefox 4. It's bar is very customizable and you can set it up to fit whatever needs you might have (ultimately resembling FF3 look, if you so please) but people still keep complaining before, during and after.
I'm not sure if GNOME is customizable or to what degree and think that this should probably be the case, BUT, what's wrong with us as users/commenters of this software/websites that we manage to make 90 % of all the feedback sound like cynical blabbering?
Disclaimer: Statistics in this post are made up to represent the poster's view.
"Science can amuse and fascinate us all, but it is engineering that changes the world. " - Asimov.
This can't possibly be the least bit confusing to Mac users who've been double-clicking in the title bar to minimize for fifteen years. /sarcasm
In general, it is safe and legal to kill your children. -- POSIX Programmer's Guide
There is a bug in GNOME which screws up the window manager if you double click the title bar.
Noted here:
https://bugzilla.gnome.org/show_bug.cgi?id=436537
"clicking in upper right corner closes window in background - very unexpected"
Very frustrating and GNOME is doing the wrong thing.
I suppose it will be good to remove the other buttons so that more people experience this, and hence a fix can be found.
liqbase
I think the Gnome developers need to check their calendars. It's the wrong month for one of these pranks.
Well, I guess there are 2 types of GUI's: ones everyone complain about and ones nobody uses.
Pretty good is actually pretty bad.
So they took a button on the screen you could click and turned into a keyboard shortcut, and one of the benefits listed in the article is that it is more touch-friendly.
It is nice that they took them out and used that space for nothing. I'm not sure how replacing useful buttons with more pixels that do nothing and convey no information helps.
Another argument given is that there's no dock or windows list to minimize to, but if you want to switch to a different window, you go to the overview, which is exactly like a windows list or dock, but less convenient.
Reading Owens explanation was painful. He starts with revealing that he never minimizes anything and then speculates randomly on why people would use it (missing nearly all of the reasons I use it), then bases everything on 2 peoples opinions who he had work without minimize buttons for a while.
The reasons for getting rid of the maximize button is they though it emphasized the title bar as a way to resize the window (WTF?) and that the new way is more enjoyable (WTFFF?)
I haven't found a single reason that wasn't based on incredibly minor aesthetics or really screwed up views of "emphasis" or "mental models."
Can anyone give an actual reason for doing this?
This sentence no verb.
... from the very beginning.
I lost track of all the "cool" but horrible ideas which made it into gnome.
- CORBA after it had long died
- XML
- GConf (the horrors of the windows registry re-implemented by monkeys)
- C# and Mono - embracing Microsoft technology!
- Umpteen window manager changes, none good enough
The sad part is that the other DE's are not in a good shape either. KDE 4 has come out of the woods recently. Enlightenment is still not out. XFCE does not have that traction. GNUstep is like HURD, barely alive.
May be writing a good GUI is beyond something that can be accomplished by a mainly volunteer community...
Nothing. It's just the latest from the Department of Stuff Nobody Asked For.
Who exactly is supposed to be the target audience for these inanities? On the one hand, you have people who have already being been using computers for a long time. They already know how to work a standard Win/Lin interface. What's the need to present a "baby" interface?
For children? 5-year olds can (and do) run current versions of GNOME without a problem.
Meanwhile, how many mod points do you want to bet that Gnome still will not have fixed 5 or 10 year old basic usability bugs in the file chooser, etc., as opposed to creating whole new ones with shiny, fancy stuff?
I'm not a lawyer, but I play one on the Internet. Blog
ROTFL.
Knowledge is power; knowledge shared is power lost.
Somewhere a million Microsoft employees are smiling.
And they're smiling because, when the boss comes in while they're looking at Natalie Portman pictures, they can click the Minimize button...
And why is constant change better *inherently* than stable interfaces?
Consider: round steering wheel, left to go left, right to go right. Brake, accelerator on pedals.
Is there a need for a car company to replace that with, say, a touchpad plus software stop/go buttons just for the sake of "Mini-Steve Jobs" points?
I'm not a lawyer, but I play one on the Internet. Blog
But,double-click-titlebar should roll-up the window
In KDE you still have the minimize and maximize buttons. "If you want to minimize, you can right-click the titlebar, then click minimize, or using ALT+F9"??? WTF, I have been using computers since 1975 and find that difficult, how about n00bs?
In KDE, if you are a "power user", you can middle-click the maximize button to maximize the window vertically while maintaining the horizontal size, or right-click it to maximize the window only horizontally. Nice, easy, simple, and keeps working what has always been working.
In other words, you can always improve something, but ***IF IT AIN'T BROKEN, DON'T FIX IT***
I paid for the [1920px wide] real estate, why would I want to waste most of it?
There's a reason that newspapers have several columns of text, and this reason is that text becomes harder to read once lines become longer than 80 characters (roughly 40em in CSS). With long lines, too much of the effort is spent on finding the next line of text and making sure you don't repeat or skip a line. So put one browser window in half the screen and the other browser window in the other half.
Sounds like you want KDE.
In KDE whether or not you want minimise or maximize buttons is a simple click in the control panel.
You only use ONE application at a time? Ok... 1980 called, it wants its workflow back.
It's not a waste if:
1) Maximizing the window makes it actually harder to absorb the content in that window (as is the case with web browsers)
2) I use the extra space for something I look at often, like my IM buddy list and messages.
Comment of the year
Do you have a tiny monitor? I could see maximizing on a laptop with a smallish resolution, but on a 1920x1280 desktop screen the concept is ridiculous.
1920x1200 here. I run most of my major and often-used programs maximized; browser, email client, Adobe Creative Suite applications, IDEs, etc. Winamp and FileZilla are some exceptions.
People will pass up steak once a week, for crap every day.
It does NOT make the UI easier to use. Cleaner looking, yes, but NOT easier. A button sits there, visible, inviting you to click it. You see that the option is exists and if you care to find out what it does you can click it and see what happens if you're adventurous or you can RTFM if you're not. Either way, you know the option exists. Double-clicking the title bar, however, is completely non-discoverable except by accident. Look at the screenshot at the top of the screen. The title bar has the title, a close button, and... NOTHING ELSE. Just a bunch of wasted space. Gnome devs are doing their users--present and future--a great disservice by removing these buttons.
I think they're trying to copy the super-clean look of iOS, but iOS looks super clean because it works differently, not because it is clean for cleanliness' sake. There is no close button because you press the home button to leave the app. There is no minimize/restore because that's not how iOS apps work. There are no scroll bars because you scroll by dragging anywhere. Steve didn't just say "I'm going to throw away all these controls," he said "I'm going to change the UI" and as a result of THAT those controls were no longer needed. Gnome has not changed its underpinnings--it just threw away all those controls.
Double-clicking the title bar to change the window is a great shortcut for power users who know it's there because it's a nice big target and sometimes it's easier to double-click a part of the screen close to where the mouse is, rather than going after a button. But that shouldn't be the ONLY way.
Decades ago, as a kid, I absolutely HATED the original Mario games on the NES because there was all this totally undiscoverable crap where you had to jump in just the right spot to mash your head into an invisible block to get points. I thought it was the dumbest thing in the world--how could you possibly know to do that? I didn't think it was a good way to make a game back then and I'm positive that it's not a good way to make a UI now. Gnome devs are ON CRACK if they think this is a good idea.
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
Do you have a tiny monitor? I could see maximizing on a laptop with a smallish resolution, but on a 1920x1280 desktop screen the concept is ridiculous.
Not ridiculous at all. I run spreadsheets every day that use every inch of screen real estate I have and I could use more. My monitor is a 24" with 1920x1280 resolution. On my home rig where I have multiple monitors I normally keep one application maximized on each monitor. I use lots of applications that maximizing is beneficial for the way I work.
I acknowledge that some applications benefit from more horizontal real estate in a single window. I was merely pointing out that word processors and web browsers usually aren't among them.
Spreadsheets, graphics and diagrams are one thing that benefit from maximal area. However, for other applications I maximise them only vertically, if any. I have configured title-bar double-click to expand window vertically.
Text editors, terminals, email message panels run 80 character wide as they have always been. A browser window where I type this is a bit wide, 1175 px, but it still leaves room for window below be that much visible I would notice if it scrolls. If web page is too wide for it, then C-- few times will make it behave.
On 7" netbook using Ubuntu 10.10 most windows are maximised and there the usage is just fine - expect for those dialogs that do not fit on screen. But if screen size is 1280x800 or better I very seldom maximize windows.
I wonder what kind of studies they've done before writing such cases then. UI Design is something you develop from test cases, real world experience and is a research branch much more closer to psychology, anthropology, Comparative Linguistics than Computer Science. It's extremely subjective, as there isn't something like the perfect "UI", like there isn't any perfect human language or the perfect blueprint for an house. The idea that an UI be done using some kind of theoretical , mathematical-like approach just because it's done "on a computer", would be not much different from trying to develop a C++ GUI framework using a manual of anthropology just because it's "Graphical" . Hope it's not the kind of reasoning behind such decisions with GNOME 3.
Ya know what would be REALLY bold? Replace the keyboard.
A radical departure would be a wall hanging with little pictures on it. Now, the unimaginitive would likely want the pictures to represent actions and things a computer would do. NO. That's too microsoft. A BOLD wall-hanging device would use pictures that DON'T correspond to actions. Like, a cartoon of a dog is what you hit to delete files.
But here's the fun part - there are TWO pictures of dogs, both hidden behind blank tiles. When you uncover one, it shows the dog for a few seconds and then goes blank. You have to remember where that is and then find the SECOND dog to complete the action.
I know what you're thinking- "That's not innovative, I've played that game before!" oh no, not THIS bold game you haven't. THIS one changes the position of all of the pictures under the blank tiles every time you open one. So you CAN'T memorize where the 1st dog was to find the second one.
Plus you can't use fingers. Nope. You use BEANBAGS. A sensor makes sure you're at least ten feet from the control, and you have to peg the tiles in the right order, throwing blindly, to complete the action. And these are not just ANY beanbags, they're special RFID beanbags. Don't lose any, though, they are expensive and demand is outstripping production so the wait for replacements is about 6 months.
Studies have shown that people LOVE bean bag toss games. This will make the computer using experience MUCH more enjoyable.
We also have a contract with Adobe. They have hired us to make Photoshop more user friendly. Currently we're looking at the bold move of replacing the entire Photoshop paradigm with something entirely different. The final decision hasn't been made. but the current favorite concept is to replace the entire Photoshop Suite with MKV files of Underdog cartoons. (MKV, not AVI. MKV is more fresh and innovative!)
Our internal studies show people in an office would rather watch Underdog than do photo editing anyway.
This space available.
I've added buttons for:
When I get around to it I might add a button to move the window and the cursor to the other/next monitor. I also have complete control over the right-click titlebar menu which I've heavily modified.
We don't see the world as it is, we see it as we are.
-- Anais Nin
Yeah, I'm looking at you Adobe Reader print dialog! >:-|
Maximize all your windows by default, that's what I do.
I got rid of the minimize and maximize buttons a long time ago.
All windows (except for dialogs) are started maximized.
If I don't want a window maximized I double-click the title bar.
I've bound "minimize" to mouse-wheel down on the titlebar.
This way I can scroll through my windows by positioning my mouse on the titlebar and scrolling down. (remember, all windows are maximized, so the title bars are perfectly aligned).
Dragging UI elements used to suck back in the days when you had to use those touchpads to navigate around the screen. Those buttons that you only had to click once were a great kludge for that situation, but now that people finally have computers so big they fill up a desk, with a device that sits on the desk for clicking stuff, this will totally make ergonomic sense!
MSIE: The world's most standards-complaint web browser.
The Gnome developers have been removing features with every release for over a decade, this is just the next step. Give them a few more years and they will remove all interaction and running Gnome will be just like running X with no window manager.
Someone from outside (Ubuntu in this case) is applying pressure for us to change/move something to emulate Apple, so we'll remove most of that feature instead.
It's called taking your ball and going home. Ha! That'll teach em!
KDE would do the opposite: Make the feature user-relocatable and put little black arrows on each widget turning them into drop-down menus. They would also add a new section to the browsing tree in Konqueror allowing you to browse windows that are minimized/maximized and to save 'favorites' for the windows you minimize/maximize the most. This, in turn, would send people who try KDE because 'its more like Windows' running and screaming away from the environment days or weeks earlier than they had been before.
Don't assume that multiple monitors means someone is trying to juggle multiple tasks. 3 monitors has become a part of my life at work. Here's how:
Left - Maximized browser window looking at my web app
Center - Maximized IDE to develop/debug my web app, providing sufficient space for my code, log monitoring, and package browsing
Right - IM Window, Resource Monitor (particularly CPU and Memory), and a handful of widgets. If there's a web-share meeting, it gets maximized here.
When I switch to a portable environment or lose a monitor, believe me, my productivity suffers.
Whereas KDE policy is "If you disKover some empty spaKe, add an useless feature or somethinK very very irritatinK. The iKon must be shiny, rotatinK, and Kontain at least one K.", the GNOME policy is the opposite: "If you find a feature, it might confuse a user, so remove it."
The alpha 3.0 release, Project Topaz, will be the perfect GNOME's desktop, as it will have absolutely Gno features at all. It will simply use excessive amounts of system resources, and do Gnothing but sit there. This final version will contain only a single button. When the user pushes it, it pops up a beautifully anti-aliased text box on a white screen telling the user to use a pen and a piece of paper to do their work and to shut their computer off.
GNOME 2.30 will be renamed to 3.0 because it will require 3GB of RAM and a modern graphics card with OpenGL 3.0 support; the graphical debugger requires a 128-bit processor, which has Gnot yet been invented, and a 3GB video card with optional 5-D rendering capability.
GNOME's logo is a huge footprint, but it is Gnot clearly established whether it is a huge memory footprint or a huge disk footprint.
(from Uncyclopedia)
http://rocknerd.co.uk
RTFA. It's not a default, the buttons are removed for good and you can't put them back.
Really? 'Cause last I checked (which admittedly was a while ago), Gnome had a default window manager, not a required one. Has that actually changed?
I'm also not sure how you can call a keyboard shortcut "harder to use". Harder to learn, perhaps, but I would have expected someone on slashdot to understand the difference between "easy to use" and "easy to learn", even if so many so-called "UI experts" can't. The concepts are orthogonal, but in this case, something that readily involves muscle-memory (like a keyboard shortcut) is invariably going to be easier to use--and almost as invariably harder to learn.
I rarely maximize and frequently minimize windows. Rather than remove these buttons I'd rather see an additional "a maximize vertically" button that doesn't change the left and right boundaries of a window. I'm hoping that panels, and having a window list in a panel are still allowed, and that clicking on a window button in the list will alternately minimize and restore it.
I'll reserve judgment until forced to upgrade during as OS upgrade. But if I don't like it I have no problems with changing to something better. Remember the good old days when window managers would allow you to customize what would be in your title bar?
Support SETI@home
Oh jesus effing Christ. I'm not a Linux hater but by god KDE, Gnome and Ubuntu seem to be having great fun in a spectacular race to the bottom.
Minimize - in GNOME 3 default shell is GNOME Shell, and in this shell there is nowhere to minimize to. You switch between windows using either Alt+Tab, or Expose, which can be reached just moving cursor to left top corner.
As a brother poster just put, what happens when you want a window hidden? I like this window existing, I don't want to see it - click, gone. Windows does this. Mac does this. Even bloody WindowMaker, blackbox and virtually every other WM in history do this. But GNOME want to differentiate themselves by removing a feature nobody wants removed. Good for them. They can suck somewhere else.
Maximize is more tricky, but more or less I always have wanted that confusing button gone. Nevermind that it toogles between maximize/restore functionality, double click on title of the window will do much better.
"Confusing"? Confused, by a maximise button? You click it, the window fills the screen. If it's filled the screen it shrinks it back. This is elementary. By all means complain about Apple's Zoom button (which is confusing to most users who don't understand what was going through Apple's head when they put it there, which is most of them) but a maximise button is probably the simplest thing there is. And by all means put Aero Snap functionality in - it's a feature of Win7 I simply cannot do without without going crazy now - but don't take a reasonable option away from me.
These changes are risky, but I would definitely not call them rushed or stupid or "just because authors want it that way". Keep in mind, that those are hired professionals which have brought us GNOME 2.x series.
They're risky because they're absolutely retarded. You would have to be absolutely insane to fuck around with something that is, to most people, an intrinsic part of using a computer. These changes are rushed, and whether their architects brought us GNOME 2 (which is hardly a case study in excellent UI idea, all things considered) is completely irrelevant. It's a silly idea.
Before criticizing understand that GNOME 3 and Unity (which also have got lot of "love" from Slashdot flamers) is created with future controls in mind - multitouch, gestures, etc.
Windows 7 does this right. I actually thought about this today - it looks like a desktop OS, it works like a desktop OS, but it would work equally well as a touchscreen OS, and neither gets disadvantaged by it. All of the controls and buttons are touch friendly without the user even realising it. To contrast however, GNOME's approach, from the looks of things (and Unity's) is "fuck you, we want it touchscreen friendly and desktop users can lump it". Bear in mind that touchscreens are still niche devices, especially on the desktop. It's stupid, stupid, stupid.
I write bullshit
That doesn't mean it makes sense to maximize everything when you've got a monitor with a resolution of 1920x1080 or higher.
There are plenty of reasons to use applications maximized. Here are 10 off the top of my head:
1. tail -f /var/log/apache2/error_log - saves using the scrollback buffer, especially when there are lots of errors, long query strings, etc.,
2 code editing using a real tab instead of 2 or 4 spaces, and an editor that supports multiple views into a file, and multiple windows inside the editor itself. I've sometimes stretched this across TWO 1920x1200 screens;
3 web sites that aren't designed by morons back in the previous century to have a fixed
4 spreadsheets
5 graphical work, esp. on images zoomed in at 800% or more
6 various IDEs (such as eclipse)
7 games
8 file browsers (esp. with previews)
9 PIMs (not everyone uses gmail or hotmail or another webmail program)
10 table dumps
The only truths I've found are that the more screen space you have, the more uses you'll find for it, and that you eventually always want one more screen,
Not saying that I like the idea for Gnome Shell but they are moving away from the old desktop idea. Storing things on your desktop by the way has LONG been a BAD thing done only by the terminally disorganized.
For files, there are places like Documents, and program links should go in the menu. There shouldn't be anything on your desktop to look at.
I know, not how we are used to doing things, but change is not something to be feared.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
I've been running Ubuntu Netbook Remix for over a year, and until this story was posted, I had not realized the minimize and maximize buttons were gone. Everything is maximized by default (although I turned that off), and if I want to see the desktop, I click the Ubuntu icon in the upper-left. I'm freaked out by this, actually; the thought, "But where are the min/max buttons?!" had never crossed my mind until this point.
actually, i just tested live image. and boy... it rocks like hell. as much as i was skeptical about minimize, maximize removal... in 1 minute i started wondering why were they there in the first place. managing maximization like in gnome 3 fells so much more natural. and minimize? when you have smart desktop managing, there is actually no point in having it. and automatic desktops... ROCK
so... in 1 minute it felt natural
in 5, i started wondering how dumb interfacing with desktop was before shell
now, just give me good session manager and i'm willing to forget every single pain with interfaces as they were. don't know if i like docked dialogs copied from apple though, but fortunately they can easily be set off to standard behavior with settings manager. although i don't plan to do that first second, i want to be sure i have it right.
p.s. the version that came with fedora 14 annoyed the hell out of me and i went back to standard after 1 day
Signature Pro version 1.13.2-3 release 83.5 beta3try7 after-breakfast edition
...what they're doing.
If they did know, then they'd realize A) a GUI spec is pointless unless it forms part and parcel of a holistic OS platform, otherwise they might as well be trying to reinvent HTML; and B) being a *nix coding geek imparts a natural INferiority when it comes to GUI expertise, not the superiority that most of them obviously feel; and C) Personal Computing is a consumer culture with certain basic use cases and expectations that must be fulfilled, so you shouldn't be surprised that putting 'candy' on everything doesn't work when you expect people to operate their computers in some profoundly different ways.
What are those profound differences? Here's a few:
1) Leaving users to grope in the dark WRT hardware compatability, instead of marketing your software to hardware vendors by offering a simple test suite and standard, trademarked icon that shoppers can readily identify on the package. Leaving it to each distro to define hardware compatibility lists was wrong: They all sucked and were half-hearted at best. HCLs should be the Linux Foundation's job because hardware compatibility is the kernel's role.
2) Leaving budding programmers and power users without an SDK or standard IDE that allows anyone to get their feet wet and share their work with confidence (as in, it will actually run on another novice's machine instead of going down in a dependency flames). If you think this is stupid or off the mark, consider that Linux is doing really well on handhelds and both Google and the Linux Foundation have their own SDKs. No one will do a Desktop SDK because of the old-hacker politics involved and their loathing of vertical integration; LSB does not go far enough and doesn't even define a way to install software packages (all it has is the package format, but no procedures or interfaces are defined).
3) Leaving users to fight-it-out with their device settings. There are still some influential (old people) who behave like Linux video was good enough with VGA framebuffer support and /dev/dsp output for one audio app at a time. Yet others treat video and audio as simplistic and beneath their concern. This has lead, for example, to subsystems like X11 that could not support the use case of 'Change the display to these new parameters and if the user indicates they work, save those setiings'. Instead we got a situation where every distro had to write their own display settings code, and they all did it badly because the assumption that display settings were just too 'simple' for X11 itself to manage them just wasn't true.
Also, what most PC programmers and techs refer to as 'OS components' (libraries, services, etc) are astoundingly referred to as 'applications' in the Linux world. This distorts the way Linux techs relay help and tips to novice users to the point where the distinction between OS and application tends to disappear.
4) Relating to the "platform" primarily by its Kernel, a piece of software that is formless/invisible to most non-programmers. Suffice it to say that if Google were marketing a handheld "Linux" to phone users, their offering wouldn't be a tenth as successful as Android and there would be all kinds of negative politics involved that called for Gnome and KDE versions just for starters. The whole community is guilty of this misstep, which amounts to a sort of mass geek delusion. Note that Firefox didn't play this game and it succeeded because people knew how it looked and behaved by default, and any third parties changing the Firefox code were forced to change the name of their offering to something other than 'Firefox'. OTOH, "Linux" defines an almost formless sea of non-kernel alterations that we geeks expect users to become familiar with.
5) Inserting the OS people between the user and the app authors, ensuring that only the biggest enthusiasts and coder-types make an effort to interact directly with the authors. This is part of what I call 'distro culture' which itself has many ill effects. Contrast this with the App Store concept where authors upload their wares themselves, and get a communication channel to/from users.
The GNOME developers clearly don't surf for porn or they don't do it in an environment where they could get caught :)
It's like there's a unified anti-porn conspiracy. First Ubuntu makes me lose the ability to quickly cube rotate to another workspace, now GNOME prevents me from quickly minimizing. I hope they at least retain the ability to set the mouse scroll-wheel on the titlebar to shade windows! :)
1. CORBA wasn't dead when GNOME 2 was created, t.i. in 2001. It died few years later, sure, but guess what, D-BUS was created and CORBA was deprecatated;
2. What's wrong with XML?
3. You really don't know what GConf is, do you? It is nothing like binary spagetti of Windows registry. In fact, it's the best thing GNOME has delivered and afaik dconf - next generation of XML registry - will be adapted by KDE too. Now you can start running and screaming;
4. C# and Mono - and while you refuse to embrase it, huge part of desktop apps are written in C#. Good luck being pure. Also I would like to note that Tomboy is only app written in Mono which is in official GNOME stack. The rest is C and Python;
5. Don't wanna clarify?
user@ubuntubox:~$ stfu This server is going down for shutdown NOW!
so I'll reserve judgement on removing those two buttons until I've played with it.
One thing that strikes me is that a title bar is a lot of screen real estate to dedicate to just a close button.
Andrew
So now we have a title bar which is completely blank, a menu bar which is mostly blank, and a button bar which is also mostly blank. I'm going to need another monitor just to hold all the blank space.
Not that this wasn't entirely predictable.
1. Wait a minute - so it is wrong to try different just because your subconscious reacts angrily to changes that will require to retrain your motor skills? :) To be more serious, well, it's their decision. I'm fine with this decision and will try GNOME 3 out when it will be finally released. No one forces you to migrate or use it. I don't need minimize so frequently, and I can try to live without it. If you can't - fine, but don't call me idiot just because I and GNOME 3 devs want to try new paradigm;
2. You haven't done proper user support, have you? Then you would know that it is very hard to understand when pressing button will maximize and when restore it to "normal" version. this concept has been always broken, I just don't fully agree with GNOME 3 devs that removing it would solve a problem. But it should be changed;
3. Again, please, don't call me or them idiots or retarded just because you don't agree with us. It is very simple thing to ask, no?
4. It works like Windows, looks like Windows, but it is not definite desktop OS per se. Because no one is. And please, tell me it's a joke that Windows 7 can be used as touchscreen OS. Because it simply can't.
I would also like to remind you that similar anger outbursts where registered here, on Slashdot just after Canonical moved buttons from left to right. I won't dive into details why such angry flame wars shows up after such decisions, but something can be said - my truth is mine, and your truth is yours.
user@ubuntubox:~$ stfu This server is going down for shutdown NOW!
* Corba
Well, that was maybe not such a good decision. The intention of using Corba was to get a programming language agnostic communication framework for the desktop, and at the time the first experiments with Gnome were made (we're talking about the late '90), Corba was hyped for exactly that purpose. In hindsight Corba was just to complicated and unwieldy to use. Yet the Gnome team learned from their mistake. The shortcomings of Corba in plain view, the need for a more simple and extensible framework led to the creation of D-Bus. And D-Bus is nowadays even empowering KDE, creating a first class bridge between the two frameworks.
* XML
I fail to see how the Gnome devs can be made responsible for the creation of XML. Gnome was started about thee years after the first release of the first XML specification. Also, XML was created by Tim Bray et al., non of which have anything to to with the inception of Gnome.
* GConf
The horrors of the Windows registration lye in its binary format, its inaccessibility, its cryptic structure and keys as well as the fact that it tries to do multiple jobs at once. The only way in which GConf could be compared to the Windows registry is in that they both store configuration settings and that they both work with hierarchical namespaces. But GConf and the Windows registry differ in very important ways; GConf's configuration files are human readable and can be manipulated by the unix command line tool tools. GConf even comes with nice command line tools itself, so scripting is easily achieved. Try to do that with the Windows registry.
* C# and Mono
C# is a Microsoft product, so I fail to see how Gnome ought to be at fault here. Mono also was not created by an active Gnome dev (although Miguel de Icaza is a founding member of Gnome), nor was is a Gnome foundation decision to create Mono. That said, although I tend to avoid Mono applications, it offers a very slick programming environment.
* Umpteen window manager changes, none good enough
Gnome changed window managers from Enlightenment to Sawfish, from Sawfish to Metacity. The next planned step, Mutter, is just a branch of Metacity relying upon the Clutter library. There were very sound reasons for those changes; Enlightenment was its own desktop project and used a different toolkit. Sawfish was written in Scheme, which non of the developers was willing to maintain anymore. As for 'none good enough', that is oppinion.
You fail to see the reasons for dropping those buttons; the Gnome developers are willing to innovate and go beyond the desktop paradigms of the last 25 years. Most of the GUI interaction concepts used today stem from those early years and even many of them provide adequate, some are just plain wrong and even hurtful. We may be using them everyday and thus they seem to be working, but many times we're just working around them. I could name many problems but they mosty center around the theme 'force me to manage my applications instead of letting my do my actual work'. Do you're research and you'll find many topics.
The current drive in the OpenSource community to innovate in desktop paradigms (KDE 4, Unity, gnome-shell) shows that this is a real hurt and I hope that the most obvious flaws of the current GUI metaphors will be addressed soon.
Try to be open and accept that the ways we've been doing things in the last 25 years may not be the final answers to graphical computer interaction. We've gathered a lot of experience, maybe it's a good time to try out some new things.
This is probably a troll, but you will find on Windows that you cannot use ctrl+c/v to cut and paste in the terminal. You use ctrl+shift+c/v. Amazingly enough, this is the EXACT SAME keys you use in most terminal emulators on Linux (the ones you seem to think require the context menu, without realizing that Windows has the exact same context menu in it's terminal!)
I also find it hilarious that you say "In windows ctrl-v or shift-ins is the universal way to copy/paste" and then later say "Linux paste is schizophrenic: ctrl-v, shift-ins". You then try to make Linux sound more confusing by adding "shift-ctrl-ins" (which exists on Windows) and the context menu paste (which exists on Windows in the same terminal emulator programs). You could at least have mentioned Emacs which uses different keys (but of course you can run Emacs on Windows and it has the SAME keys).
I was wondering about this too. Getting rid of the close box was one of the earlier changes tried by window managers (wm2 for instance). Even 20 years ago most programs had a way to close them other than the close box. The only counter examples I can think of were trivial demo programs written without any toolkit but making them quit on a the user hitting Escape would require very little code added to the demo.
It does seem like the titlebar can be eliminated entirely. All operations could be done by the user clicking in "dead" areas of the application. If you click on the blank area in your control panel, you can move the window around, or if you are near enough to the edge you resize it (probably like most Linux WMs do now, you can in effect make the entire window dead area and draggable by holding down Alt or something). The maximize, minimize, and close have been removed. The text in the title bar is either useless or repeated inside the window somewhere. So the title bar is no longer needed.
This would also get rid of the annoying distinction between "maximized" and "full screen".
... from the very beginning.
I lost track of all the "cool" but horrible ideas which made it into gnome.
/---/ /---/
- GConf (the horrors of the windows registry re-implemented by monkeys)
- C# and Mono - embracing Microsoft technology!
GConf have only superficial similarities with the MS Windows Registry. It is more similar to Mac OS X plists (they both use xml-files and deamons that report changes in them, the most important difference is that GConf actually works) and good old fashion unix configuration files. Actually, it is good old fashion unix configuration files, but in xml-format and with a deamon that alert associated application when one of them change. GConf also has an editor to make changes in those files (superficially resembling the Windows registry editor that is used to edit the the MS Windows registry database (one very large file)) and a set of command line tools, if you don't like those tools you can use any text editor you want to change your settings (look in the ~/.gconf directory, it is ... gasp!... full of plain text files).
The Windows registry is a really, really bad idea that has gone far to long. A big and fragile blob of a database that crashes everything once corrupted. To liken gconf with the Windows registry is not fair at all, if you like to compare it with anything, compare it with Mac OS X:s plists, both systems consists of many small separate xml configuration files (only plists are ususally larger and sometimes clash with other configuration files in unexpected ways, it also has (undocumented) deamons that take values from the plists and transfer them to parts of the system that use "classic" unix configuration files, but those parts of OS X become less with time, there used to be many of them but the only one I can think of that is left in 10.6 is CUPS, the time space between a change in a plist and in the other config file used to be a huge source of crashes in Mac OS X). I bet you will find the comparision favorable to GConf.
That said, XML is not very human friendly. They could have picked a simpler to edit/read file format in GConf.
Mono seem like a very bad idea. In my experience, the Mono platform encourage application makers to make really horrible user interfaces and when I have to run a Mono-application, even for the simplest of tasks, my otherwise cool and silent computer is transformed to a very noisy space heater. Most script languages produce applications that run faster and the crash frequency is just horrible.
You know, when they announced GConf, I thought like you, because I had bad experiences with the registry and back then I was an arrogant 20 year old programmer who didn't like any framework for anything, especially not tasks like configuration. However, I have warmed to it as Gnome 2.0 slowly became an OK system. Configuration should not be something that needs to be re-designed and re-implemented in every application. Things like policies, handling multiple instances of the one program, external configuration tools, etc. should work on a layer below the app. When I am writing an application, I want to be able to say, "OK, I have these options, make sure they are stored somewhere and tell me when the user wants to change them" and it works really well. It's no harder to use than libconfig or libxml2 but it has the addition that it notifies the application when a variable changes. It is a shitload easier than writing a custom parser by hand or with bison.
When Argumentum ad Hominem falls short, try Argumentum ad Matrem
"Next up we're removing the top border completely, you don't get to move your windows around the desktop, we will decide where your windows are going to be. You might not like it at first, but trust us, this will allow for a consistent level of quality as to how the windows are positioned and take up less space! Damn we're brilliant. What's that? Year of the Linux? Maybe next century when our potential users no longer need monitors or a keyboard and mouse. Then we can decide what gets downloaded into their brains!"
"we've got trenchcoats and bad attitudes" - John Constantine, HellBlazer
One of the problems with the Weird Innovation Syndrome afflicting GNOME and Ubuntu, and the reason for the intensive reaction from the community, is that there hardly ever a coherent reason for the changes.
Take the dropping of min/max buttons: Who is this for? For advanced users? Were they hobbled by existence of the buttons?
Or is it for newbs? Well, how is a noob supposed to know (and have memorized) the keyboard combinations? It used to be that the keyboard combos were nicely listed in the window control menu accessible by a button on the left top corner, but, conveniently, they're dropping that too.
What happened to discoverability, and allowing users to have simple, easy, direct methods of operation instead of forcing them to become power users from the outset and fiddling with mouse gestures?
I'm not a lawyer, but I play one on the Internet. Blog
Hey Sal, you said:
>>UI Design is_closer to psychology, anthropology, Comparative Linguistics than Computer Science. It's **extremely subjective** as there isn't something like the perfect "UI", like there isn't any perfect human language or the perfect blueprint for an house. emphasis added
So b/c there is no 'perfect UI' then Usability can't be a 'real' science? So then where is the perfect model of the universe? Because surely by the same standard Physics would have had to produce such a document.
As a scientist, I'm sure you see the logical fallacy of your claim. It is the method, data, and analysis approach that makes research truly 'science' or more in your example of "UI Design" case studies.
Yes it is true, most Usability and Human/Computer Interaction (HCI)studies use only case reports and ordinal data. Yes typically psychology and similar disciplines use that same approach.
However, HCI researchers are becoming closer to true cyberneticists (http://www.colorado.edu/communication/meta-discourses/Papers/App_Papers/McGarry.htm) every day.
Communications, specifically telecommunications, will help us take the analog signal of human intelligence and link it with our digital creations....it is already...quantization bits
Thank you Dave Raggett
By all means change the GNOME UI - drop whatever you want - but please give it as much configurability as possble (even if these are hidden as "advanced options"). Firefox 4 makes the blunder of removing the status bar (which is fine if they want that as a default) but has no way of configuring it back without an extension (bad move). It sounds like GNOME 3 are going the same way - removing stuff and not letting users configure it back.
Also, make configurability obvious - one late-in-the-day GNOME 2 change was to make GNOME Terminal flash its cursor by default (which I hate because it's very distracting) and then remove the "flashing cursor" option from the preferences of GNOME Terminal! GNOME devs claim it was because they intro'ed a global config option for flashing carets/cursors so where is that config? It's in the *Keyboard* config tool would you believe it - nutty beyond belief.
BTW, drag to top of screen to maximise is an *appalling* UI decision 'borrowed" from Windows (whenever I'm in Windows, I keep accidentally maximising dragged windows because of this brain-dead default action). Next we'll see the "jump to top or bottom of document if you drift left or right when dragging a vertical scrollbar button" (another disastrous UI feature of Windows that I always get hit by when in Windows).