I have to admit I’ve been struggling with this one for a while, most importantly because I haven’t used GNOME 3 yet; I’m setting up my own little company right now, so time’s a little short. As such, all I can do here is cover the arguments given – which, in my book, have both their strong as well as their weak points. Yes, GNOME 3 will not have buttons for minimising or maximising windows.
GNOME 3 uses an entirely new interaction model which I, as said, have not yet experienced. This makes it difficult to properly cover this subject, and as such, I think it makes sense to focus on the reasons given, as well as list a few of my own concerns that haven’t been addressed.
Let’s focus on minimising first. The biggest reason behind removing this button is a very practical one: GNOME 3 has no place to minimise windows to. There’s no taskbar or dock, and I guess my personal favourite (iconification) hasn’t even been considered (that makes me a sad unicorn). Without such a place to minimise to, it indeed seems to be incredibly counter-intuitive to have windows minimise to… Nothing? Users would be flummoxed by disappearing windows.
The second important reason to remove the minimise button is that the GNOME 3 developers believe that workspaces and activities are sufficient and can replace the need for minimising windows. “Want to see a particular window? Go to the overview. Want things to be neat and tidy? Organise them using workspaces,” explains GNOME’s Allan Day, “A lot of work has gone into designing and implementing both the overview and the new workspaces functionality – we want to focus on the best parts of GNOME 3 rather than carrying incomplete legacy functionality. And focusing on one set of functionality results in a more streamlined user experience.”
As for the maximise button, the reasons behind that change are pretty similar to reason number two above: there’s new functionality in GNOME 3 that will replace the maximise button. GNOME 3 basically implements Windows 7’s Aero Snap (I love it when good ideas are copied – and thus why I hate the current patent system), and the developers believe the dragging/snapping functionality of the titlebar in GNOME 3 can replace the maximise button. Double-click-to-maximise still works (and I hope the feature of double-click-to-minimise which I got into GNOME through a bug report does so too). Keyboard shortcuts are still in, too.
All in all, these reasons seem somewhat sound, although there’s a couple of things they’re leaving out. First, there’s the discoverability issue: the Aero Snap-esque titlebar operations are essentially gestures, and gestures are notoriously undiscoverable. New users will have a hard time finding out they’ll need to perform gestures with the titlebar to maximise their windows.
Second, and more importantly in my book, click and drag operations are incredibly intensive. With this I mean that dragging requires a lot of muscle tension, and is incredibly uncomfortable. I think user interfaces should strive to reduce the number of drag operations – not increase them. This is my biggest worry.
However, as said, I haven’t the time right now to actually use GNOME 3, so I’ll have to reserve judgment until I do. You guys, on the other hand – don’t.
That’s the beautiful thing about open source, someone can, and most likely will, release a variant with them in place
all they need is just ship with default setting
gconftool-2 -s -t string /desktop/gnome/shell/windows/button_layout “:minimize,maximize,close”
no one removed them forever. i wonder if this nonsense from news sites is just to get more clicks on flamewars
I am worried that Gnome3 won’t suit my current workflow. I do use minimize, and I must have a visible place where applications can minimize to, a panel, a dock (preferably not on the Desktop, because I want to use my Desktop to put current documents on in iconized form).
I should try out a few of the new paradigms, but I am happy with my current way of working, and I have currently started to try out XFCE 4.8.
This was my approach. I gave Gnome 3 a few days when I loaded up Fedora 15 Alpha TC2 and quickly decided that this will not work – I even tried with Avant Window Navigator to give it a bit more of a Mac feel; still not good. Gnome 3 is unrefined, uncustomizable, and all around rough around the edges.
I quickly loaded up XFCE 4.8, clicked the option to start the gnome services, got nm-applet running in the notification area, and didn’t look back.
How difficult would it be to fork Gnome 3 and include the features they abandoned from Gnome 2 – or just ignore Gnome 3 entirely in favor of Gnome 2?
I have used gnome shell on and off. Its not too bad of a concept, but it will take a while for the tweaks necessary for it to be more useful than what it replaces.
I think they should have kept the buttons, but changed the functionality of them.
All windows have a z order that indicates where they are in the “pile” of windows on the screen. Have the minimize button decrease the z order ( moving it behind other windows), while the max button increases it ( moving it closer to the top.
Not sure about the max button still. Maybe that can go away. Just clicking on a window should raise it to the top. Not sure what the max button woudl do differently, maybe just move it up one level?
I don’t think the whole crazy excuse of ” use activities to make things neat and tidy” is a good reason. I don’t know what activities I want on a daily basis. They change as new tasks come up. Requiring me to organize them is counter productive and silly. Just like you see a bunch of “untitled.txt” files on some users desktops, you’ll see a bunch of similarly organised activities just created because they need to temporarily move the freaking window out of the way. Then they have to find which activity it was in when they have to use it again. All they wanted was a temporary “get out of my way for now ” button.
Just an FYI, in case you’re not already aware of this but middle-clicking the titlebar in most *nix WMs will drop the window to the bottom of the stack.
Unless you’ve missed the last year of Gnome3 updates you will know that Gnome-Shell is built exclusively on-top of MutterWM with no option to change the VM
Yes, I’m aware of that; what does it have to do with what I wrote?
Good point. I knew that but its not a feature I use regularly … because there isn’t a button on the title bar that does it.
It seems the people who made this decision saw the purpose disappear and eliminated the buttons, without stopping to think there were buttons there in the first place: to make it easier to work with the UI. Now that its a different paradigm, what functions could benefit from having buttons on the title bar?
I really really like GNOME 2 and I’ve been a bit worried about GNOME 3 for a while, mostrly because I don’t think GNOME-Shell is on par with the GNOME experience. That said, I’ve used GNOME-Shell on various ocassions and I can say it’s getting much better, altough not quite as good as I want it to be.
On the other hand, GNOME 3 is not GNOME-Shell alone. Many other things changer for the better. And you don’t have to use GNOME-Shell, you can still use Metacity, Mutter, Compiz, etc.. You can still have the GNOME 2 experience wrapped in GNOME 3 candy.
And the maximize and minimize are gone by default, but you can still add them back if you like.
I don’t think GNOME 3.0 will be very good, but there’s much room for improvement. I still remember the mess that was GNOME 2.0 and it took many point releases to reach the maturity that we now have.
So instead of easily recognizable buttons, they’ll just have blank space on the windows?
On what planet does this make sense?
I’m using custom FVWM config, I don’t even have title bar.
I never minimize windows, I don’t use window shading….
Instead I use virtual desktops (workspaces)
and Super_L (Win-key) + Double click in window to maximize them….
works well
http://tmrepository.com/trademarks/youdontneedthat/
Really … The attitude of “I don’t need title bars … so therefore you don’t”, isn’t exactly helpful.
I agree that removing title bars is unrealistic for the general population (I wouldn’t even do it!), but linking to TM Repository kind of takes away all your credibility…
Especially when he’s doing it all the time, using it as a replacement for proper argumentation…
But if the shoe fits? For all the talk of “community” there certainly seems to be a lot of “my way is the RIGHT way” in Linux and that TM accurately points that out.
Just because developer foo doesn’t like something doesn’t mean the rest of the world is like developer foo, that is why we have these things called focus groups where we stick actual users in front of programs and see how things work in the real world.
As many people happily sign up for beta testing don’t tell me finding out what users did via focus groups would cause money, because many would happily volunteer just to help the project. No this is just another case of THIS is NOT a democracy BOFH attitude that frankly helps no one. Remember TM repos wouldn’t exist if so many cliche bad attitude stereotypes in FOSS weren’t true.
Hey, now. I’m not sure what you’re not talking about, but it’s not “Linux”.
Are you talking about Ubuntu/Shuttleworth? Yeah. He thinks he’s awesome, and knows everything about everything.
GNOME? The devs think ‘users’ are fscktarded and have no ability to think… at all. Any options will scare them.
Much less of a problem for the more sensible distros and the KDE project(s).
Debian has a much more minimal install than Ubuntu, and doesn’t break if you change out one default element for something else (you’re kinda stuck with BASH and DASH on your system, but I’ve not encountered any severe problems with replacing other elements).
Yeah…
I^aEURTMm pretty sure that the commenter to whom you^aEURTMre replying doesn^aEURTMt need title bars.
I^aEURTMm pretty sure that the comment to which you^aEURTMre responding isn^aEURTMt suggesting that other people don^aEURTMt title bars simply because this one person doesn^aEURTMt need them. Moreover, I^aEURTMm pretty sure that the commenter wasn^aEURTMt even trying to imply that either. Rather, that commenter was responding to the inquiry ^aEURoeOn what planet does this make sense?^aEUR, and I^aEURTMm sure that the take away message from his comment is ^aEURoemine^aEUR.
Moreover, I think that the principle that you^aEURTMre invoking, while a good one, has been applied to the wrong ancestor in this threat. I^aEURTMm pretty sure that the parent commenter was implying that this made no sense to him and therefore shouldn^aEURTMt make sense to anybody. If I^aEURTMm right, then you^aEURTMre principled condemnation was ironically misplaced. If I^aEURTMm wrong, then you^aEURTMre response was altogether unwarranted. However, thank you for reminding us to be sensible.
What I was saying was, you’ll still have bars minus a button. What is the point of just having that wasted space? Make the title bar tabular if you want to go there.
really? and your attitude prevents him to express his preference? he never said everyone should switch to his setup
mine are even more unready for average user
i didn’t disable my titlebar, but i disabled whole lot of other things, including minimize.
i don’t use nautilus for desktop, just to have it clean
i don’t have applications menu, gnome-do with custom shortcut
since i don’t use minimize, i don’t need window list
i abolished panel launchers long ago
i only have top panel in right corner just to fit notification bar
this setup sure is damn fun when i see when people sit behind my computer and try to do anything, now the idea with disabling titlebar will make it even more interesting.
Whoa, whoa…
I think you’re misunderstanding the discussion.
Take a breath, then a look at your English textbook.
I’m not trying to insult you, I’m just saying I think you’re not quite understanding what either party is saying.
A few words here-and-there can cause large shifts in meaning.
Example:
“Bio?
No, thanks. Not a vegetarian”
huh? What does that mean?
Edited 2011-03-08 01:15 UTC
[q]I don’t use window shading[q/]
Heh, window shading. Was that ever used by anyone for a reason other than “hey, look! Cool, eh?”?
For the record, I use window shading daily, much more than I minimize. If I double click the title bar, the window will “roll up”. (that’s what you’r talking about, right?)
I find it nicer than minimizing. It (almost) removes the window so I can see behind it, and requires a lot less mouse movement than minimize, go to the taskbar, restore.
It’s a feature I miss when I use a user interface that doesn’t have it.
I use it to quickly glance at desktop overlays like Conky and always-below windows like desktop stickies without having to find, set up, and worry about the reliability of a hide/show desktop panel button.
(LXDE/OpenBox user here… until I can get around to finishing setting up AwesomeWM with bindings I like and pleasing floating (non-tiling) behaviours for transient dialogs. I’d probably have done it by now if AwesomeWM had –replace so I could more easily switch into it for just a brief tweaking sessing.)
Short answer: yes.
It’s useful enough that a lot of Mac users who depend on it (via WindowShade X) cried foul when that haxie briefly wasn’t supported on Snow Leopard.
I imagine the admittedly small community of WS X users are in a furor over Lion dropping Rosetta. No more haxies until Unsanity moves away from Rosetta dependency.
This is an example of Microsoft style of innovation, where entire UI is messed up to solve some imaginary problems of imaginary customers. Who was complaining about these buttons? At least one? Really? It it the biggest customer’s issue when using gnome? I understand, that they are trying to do something new, but average user do not want something new in every version. He wants familiar and convenient interface, where everything works. But fixing bugs and thinking about file-copy scenarios is borring. People want “activities”.
Firstly when was this about Microsoft … and have you used Windows 7, lots nicer to use than XP and VISTA … however this is not what the article is about.
It not about customers complaining, it about usability studies and watching how people interact with the system … for example “heat maps” are used frequently by for Websites … e.g.
http://www.useit.com/alertbox/reading_pattern.html
It about watching how people use the system and make it easier for them to users e.g. Microsoft obviously watched how people interacted with Windows and made the UI changes in 7 that are more than aestectics. It implemented some features (aero snap) that I didn’t know that I wanted.
Whether for Gnome 3 they did this is another thing.
Hardly.
Win7 is Vista! God damn, people are impressionable. MS re-jig a handful of UI elements, change the name and suddenly the most maligned OS since ME becomes the most praised.
Which is as fundamentally a retarded approach as Gnome 3’s insular and carved-in-stone vision. You don’t let the computer illiterate design your UI any more than you force all users into one particular workflow.
Snap is an abomination, up there with auto-hiding the taskbar. You don’t expose functionality (and functionality that’s been there for over a decade, FTR) by making the desktop hyper-sensitive. Most people who had a use for that rudimentary tiling already knew about it, snap just ensures that your grandparents can’t avoid stumbling upon it, despite having no use for it and it only serving to disorient them.
A far saner and more flexible approach would have been to simply augment move & resize behaviour so that windows implicitly conform to a grid, rather than moving & resizing windows pixel-by-pixel. That’s something that would expose itself unobtrusively, actually make manual window management useful and provide far more options to power users than just splitting the screen 50/50.
I use XP 32bit and 7 32 bit on the same machine and 7 manages resources far better and I spend less time making coffees while the machine catches up.
It manages my dual core processor better than on XP. I see the results everyday … because I use XP and 7 side to side 9 hours a day …
No it isn’t, This is like saying that MacOSX 10.4 and 10.5 are the same … very similar yes, but not the same.
Vista RTM and Windows 7 RTM have very different kernels. Also 7 is far more modular (We are using 2008 R2 and we have several servers which are command line only). Most of this work started in Vista.
But the myth that Vista and 7 being pretty much the same will continue … far after they have both been retired…. why don’t you actually do some research.
Yes the functionality has been there for a decade, but it is a pain in the arse to use. I have to do at least 4 or 5 clicks while holding down ctrl to tile two windows vertically (I know this because I have to use an XP machine at work) or I can just throw the window at either side of the screen … the first sounds so much easier….
The “my nan” can’t do it is pretty much a TM
oh wait it is ..
http://tmrepository.com/trademarks/linuxforgrandmas/
I had what you are suggesting existed on AutoCad R14 as a auto snap feature. It was annoying as hell because it would snap to the wrong point … I spent many hours using AutoCad at Uni, and snap was onerous at best.
50/50 split is because most people want to compare two things side by side, I do it everyday, Documentation on the left, code on the right.
Furthermore Windows 7 UI snapping is optimized for Widescreen monitors, which pretty much every laptop or desktop screen on the market today. It is done for most people, which obviously obviously doesn’t include you, but that isn’t the point.
I find it particular useful. If I need something more complicated I can still resize them myself.
Edited 2011-03-08 10:04 UTC
Whose argument were you making here?
Moderately different and in ways that are largely imperceptible to users.
We’re both making arguments by assertion and offering our own anecdotal experiences. This is just a silly bit of misdirection.
While we’re being facetious and misusing ellipses, the first also sounds so likely to trip up people who didn’t want their window tiled/maximised but made the mistake of bumping a screen edge. The traditional way of tiling isn’t exposed well but it’s also not going to negatively impact anyone by getting in the way.
When you figure out what point you were trying to make, let me know as I’m at a loss as to what you were getting at, here.
Because a 2D field of windows is comparable to a CAD programme. Hell, gridless edge snapping/edge resistance has existed in X11 WMs for years and it’s just about a complete non-issue there. Defining a persistent desktop grid would solve what few issues arise from the current gridless snapping/resistance.
I’m sure most people (the ‘most’ who maximise every window excepted) would also like to be able to keep a lot more on screen at once, as well but they’re fucked if they can do it without fiddling around with manual window management.
I’m sporting a 16:9 display. You were saying?
The irony here is this could equally be an argument against Snap. ‘If I want two windows side-by-side, I’ll put them there myself’. Of course it takes 10 minutes to line them up properly, which is why you have automatic tiling mechanisms: to get around the fact that manual window resizing and moving sucks and no one likes doing it. The difference is whether the implementation caters to more than a single workflo… wait, are you a Gnome contributor?!
Uh, no.
No more restarts after every update. Completely componetised servers. Easier kernel/low-level Windows development leading to speedier, less buggy updates. Faster kernel/low-level development. Changes in kernel no longer cause negative effects upwards of the stack.
And so on.
These are all very real benefits for users, even if they may not be able to name them specifically. Calling these minor changes or imperceivable is idiotic, at best.
I love how most of what you just mentioned is only tangentially related to the kernel.
Well, you were the one who originally said Win7 and Vista are the same thing, you didn’t mention anything about kernels.
Original lucas_maximus statement, which I explicitly quoted:
It’s right there, in the comment to which Thom responded. Disingenuous much?
Actually, YOU said “Win7 is Vista!” in the comment http://www.osnews.com/thread?465200
Disingenious much?
Yes, that’s called rhetoric and being melodramatic. It was obviously not a technical discussion at that point but hey, if you want to be pedantic:
Me:
Look! I admit that they’re not the same in the same comment, nay, paragraph that I claim they’re exactly the same. Either I’m a lumbering buffoon who goes about contradicting myself in the most blatant way possible, or it’s obviously not a literal statement.
Nevermind that it wasn’t the comment Thom even replied to, or quoted.
Euh, I think you don’t understand this at all. The kernel and the related systems have been componetised and cleaned up, and that brings the mentioned benefits. They are very much related to the kernel and related systems.
Remember, Windows NT is not monolithic, meaning the kernel in and of itself is quite small and limited; when you’re talking about Windows NT’s kernel, you generally ALSO mean the low-level systems that are closely related to it, but not technically part of the kernel.
Well I clearly wasn’t using such a nebulous definition and I doubt Lucas was either, considering he raised some of your same points independent of the ‘kernel’.
It’s not nebulous at all when talking about NT. Look at the architecture diagram:
http://en.wikipedia.org/wiki/File:Windows_2000_architecture.svg
The actual kernel is pretty small and meaningless, as is often the case with microkernel and microkernel-like systems. All the benefits that I mentioned are (partly or wholly) caused by componetising and cleaning up everything in kernelmode. This process started somewhere in 2003, and was still going running up to Windows 7’s release.
If you believe such an undertaking are small changes, then you really shouldn’t be talking about a subject like this.
So for you, the NT executive is not a part of the kernel ?
Genuine question. I know that what a kernel is is highly debatable.
Edited 2011-03-08 13:02 UTC
When I speak of the NT kernel. I pretty much mean all the elements in kernelmode, which includes the executive. However, because Windows NT has a hybrid kernel – structured as a microkernel but with most elements still running in kernelmode – its actual kernel, according to the technical definition, excludes the executive.
In NT, the actual kernel is called the microkernel, and is similar to other microkernels. However, wherea with f. ex. QNX only the microkernel itself (Neutrino) runs in kernelspace and all other systems in userspace, NT has not just its microkernel running in kernelspace, but also most of the other systems which run in userspace in QNX.
This is what allows Microsoft to occasionally move a subsystem from kernelspace to userspace if it makes sense to do so. In a strict monolithic design, you would need to decouple and rewrite the subsystem substantially, or come up with a new one altogether.
You can argue the toss all you want, but most people could “get the gist” of what I was saying even though it was not completely 100% correct.
You make a lot of noise and do the routine of pretending not to understand or you lack signicant lateral thinking … either way you wouldn’t be much fun on a night out.
At the end of the day, Win7 is very different under the hood and there are quite a few nice UI additions which make it useful for the vast majority of people I have spoken to.
The uptake on Windows 7 compared to Vista is a testament to itself.
Not to mention one of the nice things about Windows is if the parent doesn’t like AeroSnap then he can simply disable it either from within Windows itself or I prefer using the excellent Ultimate Tweaker..
http://www.thewindowsclub.com/ultimate-windows-tweaker-v2-a-tweak-u…
This tool has no install, can be dropped onto a flash, and pretty much will let you customize any behavior in Windows 7. And I have to agree about being tired of the “7 is just Vista” FUD, as I ran Vista from Beta 1 through SP1 and they were like night and day. Vista was a bloated slow crash prone mess, whereas 7 is actually faster for me than XP while being rock solid and with greater ease of use.
But the nice thing about Windows is you don’t need to dig into the guts and rewrite conf files just to change it thanks to third party tools like the above. Just check the checkbox (or uncheck for those things you don’t want) save, and that’s it.
Microsoft? Really? The desktop design of Windows hasn’t really changed since Win95. All the elements are still there, taskbar, start menu, min/max/close buttons, icons on the desktop, you can even re-enable the quicklaunch bar if you want. MS is far more conservative than KDE, Gnome, or OS X.
This change for Gnome3 is much bigger than any change MS has ever made, but feel free to continue spreading falsehoods.
Exactly. And what the changes are for? I’ve tried to use Gnome, there are multiple pain points all over, as it common for almost every open-source product. File manager needs a lot of improvements, it is almost unusable. Almost all basic scenarios with simple file operations are poor thought, all menu configuration, taskbar widgets, everything needs much more attention, fixes and improvements. But minimise-maximise buttons are so trivial thing, they used to work fine.
It is just again, common sense lost to creativity rush.
I don’t entirely agree with you, but that sounds a lot more like Apple than Microsoft.
At this point I can only think of one reason to change mostly working interfaces: touch devices.
I still want to see one interface which works good(enough) with touchdevices and with keyboard/mouse.
I haven’t seen it yet (not been looking for it either)
Didn’t Microsoft copy that from KDE or Apple? I’m pretty sure they weren’t the first to have it.
At least WinAmp has had it on their application windows for a long time. So GNOME3 is hardly copying Win7 here.
Aero Snap is new. Nobody else has it. Not that it matters. But it’s a Microsoft thing.
Actually, kde has it too. Kwin copied it from win7…
KDE has it, and I would argue they actually did it better than Microsoft for one simple reason: If you have a multi display setup, say 1280×1024 * 2, the snap feature works on 0 1280 and 2560 whereas in Windows it’s only 0 and 2560.
Microsoft was the first to do it though.
While the windows don’t snap between screen boundaries, you can still use the Win+arrows shortcuts to make a window snap on the right side of the left screen (and any other combination). I’m really not sure what would be the best approach here. On one hand, snapping on every screen edge would be more consistent. On the other hand having windows snap when trying to move them from one screen to another could get annoying.
KWin handles this quite nicely. Drag a window near a screen edge and a border appears to let you know the window will:
– maximise vertically and cover 50% of screen starting on that edge, or
– maximise to cover entire screen
With multiple monitors, the border will flicker over the right-hand 50% of left screen, then left-hand 50% of right screen, then disappear as you continue dragging.
Works very well.
While Win7 may have done it first, KWin really does it much better.
Forgot to mention that the window won’t maximise until you release the mouse button, so it’s easy to move windows around without accidentally maximising/snapping them. It’s just a simple border outline that appears, the actual window doesn’t change in any way.
This is Aero sanp.
http://www.microsoft.com/windows/windows-7/features/snap.aspx
http://www.ghacks.net/2010/08/28/kde-4-5-window-tiling/
“You^aEURTMve probably seen the advertisements for Windows 7 and the tiling windows feature. Believe it or not the very first tiling window manager was Siemens^aEURTMs RTL (created in 1988). After that came a UNIX version called tAP (The Andrew Project) for the X Windowing System. Fast-forward to now and Microsoft claims to have innovated a new feature. Soon after that KDE refines the same feature and ads it into 4.4. Of course 4.4 was still suffering from numerous bugs. Fast-forward a little bit further into now and you have 4.5 which squashes tons of bugs and offers an amazing desktop that includes a bug-free tiling experience!”
It was improved again in KDE 4.6
http://bsmith1012.blogspot.com/2011/01/changes-in-kde-46-new-window…
Not a lot of difference really.
Hmm. Actually the Andrew Project had its own windowing system, originally.
There’s been a metric fuck-tonne of 3rd party apps for Windows that have implemented Snap-like behaviour for years. Unfortunately, that’s all most of your tiling options for Windows have provided. Grid Move was probably the closest.
It is unique to Windows 7 … believe it or not Microsoft actually come up with their own ideas.
KDe4 has it. If it is copied from Windows 7 or existed prior to it I don’t know.
The first time I used Win7 RC, I went to snap the windows to fill half of the monitor because I had already done so in KDE4.
I am fairly certain it was in KDE first, and I do miss being able to maximise vertically and horizontally as well as to take the full screen.
I can not remember which WM had it, but there was something that had a “fill empty space” option for windows, which maximised them up to the borders of neighbouring windows like a TWM.
Edited 2011-03-07 23:18 UTC
It wasn’t.
http://arstechnica.com/open-source/news/2009/12/hands-on-window-man…
http://www.microsoft.com/windows/windows-7/features/snap.aspx should explain what exactly Aero Snap does. I atleast haven’t seen that in WinAmp, KDE or OSX.
As for its usability: I find it just annoying, not useful.
http://www.omgubuntu.co.uk/2009/11/aero-snap-ubuntu-linux.html
As one poster noted, it was available with Beryl and enabled by default in KDE4.4 even before Win7 was released.
Also provided was a solution to do the same thing using Compiz, and many agree it was in Linux long before Win7 had it.
Indeed, KDE documentation (http://docs.kde.org/stable/en/kdebase-workspace/kcontrol/windowbeha…) lists Snap behavior, though this seems to be more or less snapping Windows to each other like WinAmp does (though that was just a quick search) – the only thing left out there (from that page) is snap to maximize, which Beryl and others have had.
A few years ago, a couple co-worked of mine were playing around with Beryl – even without KDE4 – and I’m pretty sure the Snap-to-Maximize functionality was something they were playing with. Mind you, this was before Win7 was even released as an RC or Beta.
But did it appear on KDE before or after Microsoft had Demonstrated it though?
The Beryl toying by my co-workers was before Win7 was even be talked about. So it would have to have been before MS demonstrated it – unless they demonstrated it during Vista development and then cut it out, but I doubt it. (I think a Vista RC would have been contemporary to their toying.)
EDIT: That would have been in a released version of Beryl too. So it would have likely been in development for quite some time before, and hashed around a bit too.
Edited 2011-03-07 21:46 UTC
Double post horror, nothing to see here, please move along..
Edited 2011-03-07 21:15 UTC
I use it constantly, when I am working in a SQL Server DB, for example, and writing code in VS.NET, I can have both open and side by side. On a 24inch monitor, it’s very handy.
I left these bloated, easy-to-start-using-but-ultimately-complicated “desktop environments” years ago, and don’t miss them.
Do I expect everyone to use a tiling window manager? No.
Am I very, very happy with my StumpWM setup? Yes!
I am so split on this that I can’t seem to decide if I like this or hate it. First of all, I don’t maximize windows anymore, I have a big enough monitor with high enough resolution where maximize isn’t needed, and I hardly every minimize my windows as well as I organized them to work spaces (when I am on my Ubuntu systems). SO in that sense I can relate to what is being said about it. Guess I have to use it to really find out if I love it or hate it.
I kind of dislike this move for basically the reasons Thom listed. But I don’t really care enough that I’d use something else. Right now, I think Gnome 3 might turn out better than I was expecting.
However, I think this is just more confirmation that the future of window management is tiling. As in, windows can’t overlap. Honestly, what is the point of allowing windows to overlap? Basically, if you can’t see all of a window, it’s unimportant enough that you shouldn’t see it at all, and if you can see some portion of the desktop, that’s just wasted space that should go to content.
There are a few reasons why I think people will complain.
One is that tiling is “hard”. People think of XMonad and such. I don’t think it has to be like that. There’s no reason it can’t use title bars and buttons, etc. The only reason that tiling window managers are so hard to use is because they have been limited to the elite who are okay with using only arcane keyboard shortcuts.
The other reason is that people claim to actually use drag-able, resize-able, overlapping windows. Everyone I have ever seen either maximizes most windows and switches between them with the task bar/dock/etc., or doesn’t maximize windows, but uses them as if they were (i.e. one at a time). When people actually do use more than one window at once, they almost always are using multiple monitors, or tiling (like Aero Snap and equivalents). If this is how people use windows, logically it would make more sense to use a system that actually supports this paradigm.
Personally, this is what I would like: a Unity-like dock along the left side, one main window on the right, multiple smaller, vertically stacked windows on the left, and all apps using a toolbar/menu-bar on top, tabs on bottom, and a scrollbar on the left if applicable. Then every screen edge would be maximally utilized for Fitt’s law.
I obviously am not “most people”, but I don’t quite fit in any of those definitions. I personally use minimize and maximize buttons all the time, I use some applications in maximized mode and some applications in smaller windows so I can have 4 such windows visible, not all of them necessary the same application though.
I find Aero Snap a pain in the arse, and I have quite a strong feeling I will hate GNOME3. Sad, because I’ve been a stout GNOME-supporter all through the 2.x series! :<
I think you missed the point, which is 100% understandable if you’ve never used a tiling window manager.
How often do you _use_ the ability for windows to overlap? What good does it actually do you?
How often do you need to see _part_ of a window, as opposed to either all or none of it?
I frequently have video in a window (floating on top if possible) and a web page or chat window underneath / to the side / etc which I scroll as needed.
Edit:
That said, I do find Aero snap useful in these scenarios (video snapped to the left, text to the right, but video and most web sites are optimized for wide screens, so it’s not always as nice as overlapping windows.
Edited 2011-03-07 23:43 UTC
Now that you mention it, MPlayer IS one of the default “let this application float” exceptions in AwesomeWM’s config and another case of something I let float (usually in the corner) on top of things I’m doing.
The whole “try to do the right thing” part of mixing tiling and floating is probably one of the biggest reasons I’m still not finished setting up AwesomeWM for myself.
Quite often, actually. That’s why a tiling window manager wouldn’t work for me either.
Care to elaborate?
I can give a use case: having a file listing below another window, say a text editor where I might be entering info about the files. I could use tiling but I prefer it that way because I can see more info in the front window and not have screen space used by parts of the background window that I have no use for. Which is also the reason why I can’t use Windows OSes without a utility like taekwindow that lets me scroll the background window without bringing it to the top of the window stack.
I also often happen to have portions of FreeCommander visible when I am browsing snapshots in OopsBackup (Time Machine-like solution for Windows). Sometimes, I just want to see the last modification time for a folder or file, not the entire fullscreen window of the two-pane tabbed explorer.
Quite often, actually.
It depends on the apps, and it might be solved if UIs were designed in a different way, but the thing is lots of times I’d rather just keep a couple of lines visible of the underneath app. I can cover menus, buttons and all the stuff I’m not needing with something I actually want to see. I’m still able to give focus and scroll or provide input to the app underneath without rising it.
Different workflows I guess, but with the app UIs I’ve been given that’s the one that works the best for me.
Ah, I actually forgot about buttons, menus, and other semi-useless clutter.
I don’t use things that have those, if I can help it. I don’t get showing me things I _might_ want to see at some point.
*shrug*
Actually, one of the things I thought about which I forgot to mention is that there could be two distinct modes for windows. In my hypothetical tiled environment, the main window on the right would be like normal. The stacked windows on the left could have a stripped down interface that would be designed mainly for viewing and simple editing.
I do that frequently… overlap windows so just the bits I need from one window are visible while I work in the other.
How is this possible? I use Windows 7 every day and out of all the snaps that Windows 7 does, around 90% are exactly what I wanted it to do. If not, it’s easy to undo it, just move the window around a bit. What’s your big beef?
Umm, ok, but um, wouldn’t this mean that using an app lto manage your devices is better than mounting it on the desktop and dragging crap to it?
Agreed, which makes any “dragging” operation a “gesture”, and therefore “notoriously undiscoverable” – even more reason to handle the device management through an interface that doesn’t require much in the way of “discovery” for the user isn’t it? Dragging also requires a degree of hand-eye coordination and from over twenty years of IT support experience I can very confidently say there are many people who have absolutely none.
You can perform every operation in iTunes without having to drag a single thing, so by your definition that’s better UI design.
You can do every crazy position imaginable with a prostitute. Doesn’t make it better than having regular sex with the one you love.
In non-crazy speak: just because iTunes gets *something* right doesn’t make it a good product.
Speaking from experience?
What with the Gnome foundation and Canonical wreaking havoc with the traditional Gnome desktop, I wonder how many users might ignore the now outdated FUD about KDE4 and move on over to a shiny new KDE 4.6 desktop. A desktop where everything is where users expect it; or, where things can be easily moved to where any user might want it. While KDE 4.x is still no panacea, it’s drastically improved over the past two years and is now quite powerful, usable, and a safe harbor for those who’ve had enough of the shakeup on the “other” desktop. And of course, there’s always XFCE or LXDE or even Enlightenment for those who KDE still cannot serve. Isn’t choice wonderful?
Hallelujah, someone finally thinking outside the box a bit!
It’s been a long time since I saw a major OS have the guts to try something different than the usual taskbar + windows-with-3-buttons approach.
That said, I really think that removing the maximise button is a mistake. Although I almost never use it (I double-click the title-bar), it provides an important clue for noobs.
Looking forward to seeing in action.
I’m not sure how many people will be using GNOME 3, or for how long.
GNOME 3’s not even _released_ and they’re already dropping features.
In a year GNOME will just be a big ‘foot’ button that does what the devs think you should be doing at a given time of day, pre-set to GMT, and can only be changed with gconf-editor.
I love gnome. I’m currently using it on my laptop… but the next release is looking more and more shitty
I despise all the mockups and alphas of the gnome panel so far. Its horrible. I’m really really really unhappy with the current direction gnome is taking.
I’m going to have to change WM at some point :/
Personally I think this is a very interesting development. It basically does away with window management as we know it and I can see why that would rub many people the wrong way. I say they should go for it though. Maybe it will fail in a spectacualr way but at least they tried something new and if no-one tried nothing would ever happen. Then again, I found KDE4 interesting too, even from the beginning.
Yeah, I’m a fan, if not a user of KDE4.
The biggest changes KDE4 made were under-the-hood and initially theoretical. Workspaces vs activities was a big O_o for most people until 4.6!
A mainstream desktop environment with a tiling-capable window manager? Neat. Heck, I can’t think, off hand of any tiling WM that allows you to use the mouse to adjust the layout. Scriptable with various languages? Neat!
Plasma is really cool. A plasmoid can now be run in the panel, on the desktop, in a window, full-screen… It’s really really cool!
GNOME 3 seems more about superficial changes, trying to change how the user interacts with the computer… Gnome-shell is also a big monolithic… thing.
Never been a fan of how GNOME does things, and this is no exception.
You know, Activities is the one thing in KDE4 that I don’t get. I can switch between different setup of widgets on my desktop? Uhm, why? I guess it’s because I never put any widgets on my desktop…
It can also function as an autostart system(start activity foo. applications bar, baz and quux start and so on) automatically place certain applications into particular activities, and more.
It’s still fairly complex to set up iiuc, but getting simpler.
Plasma-netbook has been doing this since 4.4? Well, 4.5 for sure.
With plasma-netbook, all windows are maximised, including modal dialogs. There’s also no window borders or title bars, which means no minimise or maximise buttons.
On small (10″ 1024×768 for example), this works wonderfully, as it adds quite a bit of extra vertical space.
On large screens (like connecting a palmtop to a 23″ 1600×1050 LCD), it’s not horrible, but not being able to see two windows at once can be challenging.
Personally, I find adding the standard taskbar widget to the plasma-netbook interface is pretty much the perfect UI for small screens.
Will be interesting to see how GNOME3’s UI works on large screens. I can see it being useful on small screens, but large ones?
KDE4 got this right, I think, by splitting plasma up into different UIs for different uses (phone, palmtop, laptop, desktop, etc).
Then gnome will be for hobbyists to play with and not for real work.
I survived for ten years on Mac OS Classic without minimize, and there was never a real maximize either.
Never even missed it. Nowadays on Linux you can just use a new workspace if you want to “get that window outta my way!”.
Supported. It’ll probably only take Windows-ites a week or so to get used to the lack of “minimize”.
Agreed. Minimise — and IMO, overlapping windows too — are due to design deficiencies. Never used classic mac, but that window-to-tab-at-bottom feature seemed nice.
I never once missed the minimize button in NeXTStep. Left side has Maximize Square and the right side had the Close X Button. All 4 sides to scale up to what I desire. This seems to take it in a user defined scale and store last used positions approach [which NeXT always had from the beginning where Apps always remembered their prior state to ending execution].
So, an endless stream of virtual desktops is supposed to be better than a taskbar? When one window opens on top of another, now suddenly you cant minimize, you have to drag the window aside or move it to another virtual desktop? How about kde and gnome stop removing highly used features (desktop icons and maximize/minimize) just so they can be “edgy” and work on features that are -actually inovative-?
I used to like kde, but I cant even use kde4 anymore. It just seems retarted to me that they would limit the placement of icons and widgets to several draggable boxes, rather than just allowing users to put them directly on the desktop.
Now it looks like gnome is going the same way by removing the taskbar, maximize AND minimize. I guess we dont need usable. Just as long as the developers of these desktop environments can feel edgy.
Edited 2011-03-08 14:10 UTC
God, I know exactly what you mean. I have ranted about the same thing on the Opera forums just two days ago when I discovered that pinning a tab in version 11.xx now shrinks the tab’s width to just the favicon and forces all pinned tabs to the left of the tab bar… which means that pinning a tab removes the page title from my sight but also messes with the order of tabs!
Devs (and designers? there are probably designers in these teams) seem to do just what they think is good because some user happened to whine about the other way.
At least with Gnome you can put the buttons back, although it would be totally useless for the minimize button. The Opera guys offer no way to keep the tab width the way it was in 10.xx so I had to revert to the last 10.xx version.
Wholly agreed. I thought about this some more, and I honestly can’t see this as a good idea.
I am a translator, and for my business I have to manage multiple windows all the time – Aero Snap takes care of the source and target text (snap target to the left, source to the right – yay 24″ display!), but at the same time, I switch between loads of browser tabs (resources for translation), emails (to manage incoming jobs and read info on current ones), and possible reference material stuck in Word/PPT/Excel/etc. files. The idea of having to fiddle with unintuitive virtual desktops and activities and dragging just to switch between windows… Dear lord I get RSI just thinking about it.
This is going to be one hell of a clusterfcuk for those of us who actually do some goddamn *work* on our machines. I need a taskbar or dock – without it, quickly switching between windows is going to be impossible.
Edited 2011-03-08 14:15 UTC
Uhm…Alt-tab?
alt+tab is sequential and respects the z-order. You can’t gain direct access to a specific app with alt-tab. I always found alt+tab pretty close to useless.
Have you tried Gnome Shell’s implementation of Alt+Tab?
As long as you don’t release Alt the application list is visible and you can use cursor arrow keys or even the mouse to select any application and even hover over applications to see the single window instances of each application (on any workspace, those not on current one neatly separated).
No seriality in selection is really forced upon you if you navigate the list by keys or mouse: it’s basically like an auto-hiding overlay dock triggered by a key combination when it comes to application/window switching.
Now, an interesting question would be if it wouldn’t make sense to unify this switcher with the dock you get in activities, from a UI perspective…
I confess I’ve never looked at KDE4, but if it puts icons into draggable boxes I probably won’t like it. That was the one feature of Common Desktop Environment (CDE) I detested back when I was using Solaris 2.5.1. I was thrilled to switch to Gnome2 when I finally got to upgrade to Solaris 8. When I started using Linux with regularity, I was happy to continue using Gnome2.
From the description, Gnome3 will likely force me to look for a new window manager. I regularly use overlapping windows on multiple virtual desktops and iconify some windows, in part because I do not have the *luxury* of dual monitors (yes, that is still a luxury for some of us).
Define “into draggable boxes”?
If you mean the CDE app drawers, the only other DE I’ve used with that was earlier versions of XFCE (understandable, since it started off inspired by CDE). KDE uses a fairly standard start menu + task bar.
It does however have an icons-in-boxes thing in that the desktop is just a container for plasma apps – such as a folder view (essentially a file manager window). In at least some versions, the default setup gave you one of those, in icon view mode, displaying your Desktop folder. And yes, you could drag it around.
This is what I mean though. Why draggable boxes? Why cant I just put everything on the “desktop”? Draggable boxes are just an added layer that dont seem to serve any purpose.
Edited 2011-03-08 18:48 UTC
You can, just right click on your desktop and change the desktop mode to “folder view”.
For instance, Mandriva does it by default. Other distros tend to keep the “upstream” behaviour.
The purpose is to keep the desktop tidy, and to have several different boxes with different views on different folders, if you want to. But it’s just an option.
Edited 2011-03-08 18:49 UTC
Maybe this is in a newer kde4. It hasnt been there in past versions Ive tried (either that or I just didnt notice and my entire argument is retarted.)
Edited 2011-03-08 18:50 UTC
It’s definitely been here from the start, you just missed it
ALL HAIL ME! KING OF THE DUMBASSES!
I’d upvote this if I could. Not because I think you’re a dumbass, but I respect anyone willing to eat crow.
The default “desktop” in KDE4 is a plasma container that you can put any widgets into. This includes Folder View widget, which lets you create a draggable box that shows the contents of a specific folder. You can create as many of these as you need/want.
Or, you can switch the “type” of the “desktop” container to be “desktop”, which just shows a single Folder View locked to show the ~/Desktop/ folder. Same as the way the “desktop” works in Windows.
IOW, you can have it whichever way you prefer.
Actually, you’re mistaken regarding KDE & “draggable boxes”. They didn’t remove features but added many of them. (which is why some people prefer a minimalistic environment, which is reasonable as well).
There are actually several types of desktop you can select. By default, you have a plasma desktop with an applet that contains the content of the “desktop” directory (folderview applet). This way you can watch several different directories from your desktop and prevent problems like applets overlapping icons. But you can also use an ordinary desktop. Or a “grid desktop” & other kinds of desktops (6 types, each can have 8 kinds of backgrounds including dynamic ones, and as many applets as you want).
Edited 2011-03-08 18:36 UTC
Yeah, it looks like there are a few things I missed there.
But overlapping icons seems to me to be a bit of a non-issue. Thats one of those things where “So dont do that” kind of applies. Anyone putting widgets on the desktop (like in windows 7) probably wouldnt do that to begin with.
Min/Max buttons are still there in KDE (plasma shell), not sure what you are complaining about.
You can still do that, just change the type of Activity to Desktop, and voila! You have a “normal” desktop where you can only put icons.
Personally, I find the flexibility of the non-desktop version to be much nicer. I have multiple File View regions on my “desktop” so that I quickly access files in the folders I use the most often. Much nicer than “just a bunch of random icons”.
Stop blaming KDE for things that GNOME has done wrong. Especially since all the things you complain about are still available in KDE. Misplaced anger, perhaps?
as a fan of CDE, you should take note, that there is a rather new project aiming to recreate a opensource implementation of the CDE concepts
http://devio.us/~kpedersen/
I’m just not there yet.
I saw there are arguments about whether W7 or KDE invented the tiling feature. Well, I don’t know, and I don’t think it matters as good ideas are generally borrowed from somewhere else & get progressively improved. What counts is the underlying philosophy and the fact that the most people benefit from those ideas.
Of course, there has been tons of “tiling” window managers under Linux for ages, but this new simple approach make it actually usable for normal people.
There are other automatic tiling options in KDE4 (spiral / columns etc…). Very powerful, but, I guess use cases are very specific and clearly not for the regular user.
There are many ideas that seemed to be invented by Apple. for instance Quartz. But some 3D window managers used to exist, as working experiments years ago. But Apple had the opportunity to implement it cleanly for the first time. (mostly because they started from scratch, with a small array of hardware to support etc.)
We could also quote the usage of virtual desktop under the AmigaOS, a long time ago (public screens actually, which was a different yet powerful idea).
Edited 2011-03-08 18:46 UTC
Lots of description in this article about why the change was necessary. But how does it work? Does anyone have a video?
Aero snap is the worst feature in Windows 7. I hate. I don’t want a window to be maximized when I drag it to the top of the screen. I just want it to align with the top of the screen. I constantly have to fight it when I’m in Windows.
In KDE, I don’t have this problem, because you can disable it. Unfortunately, I have not been able to find out how to do so in Windows 7.
I love how you can snap windows in place in KDE (like snapping a window to the top of the screen or to line up with an adjacent window), but maximizing is another beast entirely. There’s a button for that. It’s not hard to hit. In fact, it’s likely faster than dragging the window to the top of the screen. Aero snap is horrific IMHO. I’ve never liked Gnome, so I’m not sure that I care much about what they do with Gnome 3, but the more I hear about it, the less that I want to use it.
re: “Aero snap is the worst feature in Windows 7. I hate. I don’t want a window to be maximized when I drag it to the top of the screen. I just want it to align with the top of the screen. I constantly have to fight it when I’m in Windows.”
It’s actually my favourite feature. But I agree that Microsoft should let you turn it off.
Edited 2011-03-09 02:54 UTC
I guess Google is too hard for you… There are dozens of places explaining how to turn Aero Snap off.
A good compromise is to disable the maximization (which I did under KDE 4) but keep the 50% tiling feature. Because the maximization easily occurs without your wanting it when just trying to move the window.
As most people I guess, I maximize windows vertically and not horizontally, which explains why the “top aero snap active border” gets in my way and not the borders.
Edited 2011-03-09 09:07 UTC
I have been skeptical about gnome shell, but I recently decided to give it a shot since everyone is talking about it. And I was left feeling disappointed by it.
The activity area was not nearly as distracting as I thought it would be, being able to tap the window key and get a nice overview of the desktops was nice. Having to start all my applications from that would get annoying after a while, but I prefer a Gnome Do like launcher anyway, so it wouldn’t get in my way very much.
However a lot of other stuff annoyed me a lot. I felt the lack a window list very keenly. I don’t use many application and was testing on a netbook, so it wasn’t really the lack of minimize that got me, most of my stuff is maximized with that small of a screen. It was having to go back to the damn overview screen every time I wanted to switch between a few windows.
The lack of widgets, applets and ability to really do anything to desktop itself was also a huge problem for it. This was a really really dumb move on the designers part. Its not that I can’t live without that stuff, its that I don’t want to and there are other DEs that give it to me.
I may check out Gnome shell again in a another release when someone has given it a window list or a dock, applets, a places menu, and the ability to create some goddamn desktop shortcuts.
Edited 2011-03-09 04:24 UTC