GNOME Shell 2.91.90 released
From: | Owen Taylor <otaylor-AT-redhat.com> | |
To: | gnome-shell-list-AT-gnome.org, gnome-announce-list-AT-gnome.org | |
Subject: | GNOME Shell 2.91.90 released | |
Date: | Tue, 22 Feb 2011 23:11:44 -0500 (EST) | |
Message-ID: | <62650781.90599.1298434304347.JavaMail.root@zmail07.collab.prod.int.phx2.redhat.com> | |
Archive‑link: | Article |
GNOME Shell 2.91.90 is now available at: http://ftp.gnome.org/pub/GNOME/sources/gnome-shell/2.91 0cb01d1b9cdd060883a88fab9377b8c4498ac24a967c5135e6ab6c05f598e332 gnome-shell-2.91.90.tar.bz2 20e41afe1e1528856d158663a5ff4aa08feddc66d3861ecc1ce55ea44867f991 gnome-shell-2.91.90.tar.gz This release just about concludes user interface changes anticipated before GNOME 3.0. The only significant change we expect after this release is to add a native network indicator based on NetworkManager 0.9. About GNOME Shell ================= GNOME Shell provides core user interface functions for the GNOME 3 desktop, like switching to windows and launching applications. GNOME Shell takes advantage of the capabilities of modern graphics hardware and introduces innovative user interface concepts to provide a visually attractive and easy to use experience. Tarball releases are provided largely for distributions to build packages. If you are interested in building GNOME Shell from source, we would recommend building from version control using the build script described at: http://live.gnome.org/GnomeShell Not only will that give you the very latest version of this rapidly changing project, it will be much easier than get GNOME Shell and its dependencies to build from tarballs. Changes since 2.91.6 ==================== * Workspace handling [Owen, Jakub] - Replace existing workspace controls in the overview with a vertical list of workspace thumbnails. - Change workspace orientationin the main view to vertical - Workspaces are automatically managed - empty workspaces are removed, other than the last workspace which is always empty - a new workspace is added when something is started on that workspace. - Add ability to change workspace by mousewheel scrolling over thumbnails [Sardem FF7] * Add a PolicyKit authentication agent; requests to the user for authentication from PolicyKit now show up as shell-themed dialogs. [David] (http://davidz25.blogspot.com/2011/02/gnome-3-authorizatio...) * Visual refresh [Florian] - Improve the appearance and behavior of the overview "dash" - Use larger icons in the Application browser - Improve the appearance of the top panel and round the corners of the screen [Florian] - Improve the appearance of the search entry in the overview [Florian, Ray] * Remove minimize and maximize buttons from the titlebar [Owen] (http://www.mail-archive.com/gnome-shell-list@gnome.org/ms...) * Change the options for stopping the system; Suspend is now in the menu, while Power Off... is hidden, but can be accessed by holding down Alt when browsing the user status menu [Ray] * Port telepathy integration to telepathy-glib from hand-written D-Bus code [Guillaume, Morten] * Remove the window filtering and highlighting when using the dash application menu - it was confusing and buggy rather than helpful [Adel] * Use the alt-tab switcher when <Alt>Above_Tab (alt-` typically) is pressed, and fix keybinding handling durng alt-tab. [Rui] * Message tray - Improve the expand/collapse behavior of for greater stability [Marina] - Hold notifications while the user is marked busy [Giovanni] - Group chat messages together [Hellyna] - Fix bug that resulted in missing icons for contacts without avatars [Hellyna] - Enable navigation using arrow keys between buttons in notifications [Hellyna] * Add audio feedback when scrolling over the volume status icon [Giovanni] * Add a "Show Layout" item to the input source selector menu [Giovanni, Sergei] * Use GLib application launching API, to allow us to associate windows with applications in a broader rannge of circumstances [Colin] * Unify history management between run dialog and looking glass, giving consistent behavior [Jasper] * Pass extension metadata object to extensions so they can be configured in metadata.json [Giovanni] * Support symbolic colors for legacy tray icons [Dan] * Shell Toolkit: implement ability to specify inset shadows in CSS [Florian] * Remove no-longer-useful --xephyr option from gnome-shell wrapper script [Dan] * Improve the drawing of the "box pointer" used for menus and notifications [Sardem FF7] * Memory leak fixes [Maxim] * Code cleanups [Adel, Dan, Florian, Marina] * Bug fixes [Adel, Bastien, Dan, Florian, Giovanni, Jasper, Marina, Maxim, Owen, Ray, Takao] * Build fixes [Dan, Jason, Jasper, Luca, Owen, Florian, Pierre, Thomas] * Visual tweaks [Florian, Jon McCann, Jonathan S, Luca, Marina] Contributors: Giovanni Campagna, Jason D. Clinton, Guillaume Desmottes, Maxim Ermilov, Luca Ferretti, "Sardem FF7", Takao Fujiwara, Adel Gadllah, Rui Matos, Morten Mjelva, Florian Müllner, Hellyna Ng, Bastien Nocera, Jonathan Strander, Ray Strode, Jasper St. Pierre, Owen Taylor, Sergey V. Udaltsov, Colin Walters, Dan Winship, Thomas Wood, Pierre Yager, David Zeuthen, Marina Zhurakhinskaya Design: Alan Day, William Jon McCann, Jakub Steiner Translations: Khaled Hosny [ar], Petr Kovar [cz], Jorge González, Daniel Mustieles [es], Ivar Smolin [et], Fran Diéguez [gl], Sweta Kothari [gu], Yaron Shahrabani [he], Luca Ferretti [it], Changwoo Ryu [ko], Shankar Prasad [kn], Kjartan Maraas [nb], Wouter Bolsterlee [nl], A S Alam [pa], Matej Urban?i? [sl], Abduxukur Abdurixit [ug], Daniel Korostil [uk], Aron Xu [zh_CN], Chao-Hsiung Liao [zh_HK, zh_TW] Bugs fixed: 588050 Keep overview around rather than recreating it every time 594071 Setting main menu key to <Super_L> conflicts with gnome-shell 594324 Some recent documents open with the wrong application 597859 Add Help browser as an application in the overview browsing and search 599334 installing software is not a preference 600771 Add an Input Language system status indicator / menu 603759 Window opens in strange manner 604237 Remove minimize button? 609791 right clicking on user name in upper right corner doesnt work 610818 does not run in xephyr 612548 desktop zoom like compiz extensio 617225 show message tray when inactive or returning from away 620416 Port the telepathy client to telepathy-glib 631995 Consider adding a "Monitors" system status 632595 Recorder: Switch to webm 633667 No feedback on volume change from panel 636156 Improve dash dnd launcher reordering behaviour 636370 Can't use system settings (sound applet...) 636680 'Suspend' button doesn't Do The Right Thing 637687 Agent word in "Bluetooth Agent" is weird 637745 Use new GLib application API for launching 638720 crash/hang (bluetooth) 638990 Favorites: Refine default list 639324 Error trying to run gnome-shell 639341 altTab: use keybinding actions instead of clutter keysym comparison 639428 Use large launcher icons in the application view 639468 No icon for "invisible" contact 639943 Port to TpTextChannel 640361 Minor comment bug in workspace.js 640363 Leftover method Workspace._removeSelf 640465 border color seems to affect fill for .dash-search-button {} 640583 St: drop StClickable, add some functionality to StButton 640976 Avoid passing 'source' as an argument to Notification::destroy() 640978 Make sure we only emit 'destroy' for a notification once 640996 Use thumbnails for workspace management in the overview 641060 Use proper colors for trayicons with symbolic icons 641117 Strange behavior while using Ctrl+Alt+Tab 641245 Make calendar more l10n friendly 641359 configure.ac: Require GTK+ 2.99.3 for the tray API changes 641415 New app can be used by the kbd indicator icon in order to show the current layout 641522 theme-node: Box shadows broken with prerendered materials 641533 Workspace: don't show window overlays when leaving the overview 641537 Overview search entry is to short for its text. 641538 volume control OSD and shell status indicator not synced 641605 Add libxklavier to moduleset 641677 Full-screen gnome-terminal and empty workspaces causes panel to be hidden. 641726 Make the boxpointer movement smoother for the arrow 641728 Boxpointer arrows should be "real" content centered 641809 Add mouseButtonClicked argument to the callback for St.Button 'clicked' signal in Notification 641810 MessageTray: factor out focus grabbing from Notification into a separate class 641879 Shrink workspace thumbnails to prevent overflow 641880 Add a fake panel to workspace thumbnails 641881 Slide workspaces thumbnails in/out 641886 Volume should go up to 150% 641887 Shortcuts to switch workspaces are inconsistent 641896 gnome-shell does not forward GDK_KEY_PRESS to clutter_event_put 641931 workspace-switcher: Switch to vertical orientation 641973 Make clicking on the active workspace thumbnail go to the main view 641977 new workspace selector looks strange 641987 Make the app view more predictable 642005 MessageTray: fix showing and hiding summary notifications when summary items are clicked 642031 Adapting boxpointer to allow more precise placement of the arrow and match the calendar mockup 642034 Use magnifier schemas from gsettings-desktop-schemas 642058 lookingGlass: fix red border drawing in the inspector, port to JS 642059 kill off shell-drawing 642117 dnd: Fix bug in computation of snap-back position 642124 Fix fuzziness for application icons 642175 Opacity for magnifier crosshairs is now a float 642189 Remove window filtering for dash item right click menu 642192 Correct annotations for st_theme_node_get_content_box() 642194 Update calendar font styles to match mockups 642196 Split the overview's constructor into an internal and a more public part 642207 placeDisplay: Fix typo preventing places from being launched 642208 search: return an empty list instead of null 642209 status messages show markup 642237 Clean up history management between runDialog and lookingGlass. 642287 Change the search box text 642295 StThemeNode: use (out caller-allocates) on ClutterColor-returning methods 642303 workspace-thumbnails: add the ability to scroll over workspaces 642329 Minor workspace fixes 642333 st-entry: Add API to set an actor as primary/secondary icon 642334 st: Add support for inset box-shadows 642335 search-entry: Update style 642483 StEntry: remove special redundant hover tracking 642510 StScrollViewFade: Make fade-offset a property rather than a hardocded constant 642600 Need to update the accessibility gsetting variable 642641 "High Contrast" won't work when starting with "High Contrast" theme 642672 Fix workspace thumbnails in RTL locales 642697 Update panel style 642699 Segfault when closing some applications 642721 Two minor fixes for RTL locales 642726 StWidget: unset hover when unmapped 642834 Bouncing workspace selector 642886 PolicyKit authentication agent 642925 workspaceThumbnail: Remove tweens when the actor is destroyed _______________________________________________ gnome-announce-list mailing list gnome-announce-list@gnome.org http://mail.gnome.org/mailman/listinfo/gnome-announce-list
Posted Feb 23, 2011 20:25 UTC (Wed)
by arjan (subscriber, #36785)
[Link] (8 responses)
As one of the few developers that try to use MeeGo on a daily basis, I can say that it is absolutely horrible and the worst decision made in the MeeGo UI design.
I sure hope GNOME isn't going down the same rathole, and that the GNOME UI team considers "open source contributor" as one of their primary target users (in addition to the traditional "grandma" kind of user)...
open source projects thrive by (potential) contributors using their software and wanting to improve some aspects of it.. either as a one of contribution or by then getting roped into making the whole thing better in other areas too....
Posted Feb 23, 2011 20:34 UTC (Wed)
by halla (subscriber, #14185)
[Link] (1 responses)
Posted Feb 23, 2011 21:42 UTC (Wed)
by nedrichards (subscriber, #23295)
[Link]
And yes, I know that the touchpad on the S10-3t actively doesn't work (as opposed to just not work very well) but it's indicative, in our measurement and experience the target areas that a user can hit with a typical netbook touchpad are roughly equivalent to those you can hit with your finger.
Posted Feb 23, 2011 21:19 UTC (Wed)
by walters (subscriber, #7396)
[Link] (1 responses)
As far as appealing to contributors - definitely. Can you elaborate on "absolutely horrible"?
Posted Feb 23, 2011 21:51 UTC (Wed)
by nedrichards (subscriber, #23295)
[Link]
There were a number of reasons for that relating to consistency, design direction and implementation details of the window switcher. Reasons very much like the GNOME shell team in fact.
I wish we'd been able to explore some tiled window strategies as well, to try and meet the 'side by side' usage scenarios that we know exist and that Owen mentions but sadly, we never really prioritised it high enough as the side by side experience was just fundamentally unsatisfactory on 1024x600 at <10".
Posted Feb 23, 2011 21:33 UTC (Wed)
by alexl (subscriber, #19068)
[Link] (2 responses)
Posted Feb 23, 2011 21:38 UTC (Wed)
by xav (guest, #18536)
[Link] (1 responses)
Posted Feb 23, 2011 21:59 UTC (Wed)
by AlexHudson (guest, #41828)
[Link]
It makes much more sense not to have a minimization button in that scenario, but then you have to make sure you're not taking workflow away from the user. If you are, then maybe you think about some other system.
Having a window bar at the bottom just to support minimization is a bit of a waste of screen real estate.
Posted Feb 24, 2011 6:41 UTC (Thu)
by eduperez (guest, #11232)
[Link]
Posted Feb 23, 2011 21:18 UTC (Wed)
by GhePeU (subscriber, #56133)
[Link] (24 responses)
I'm using a computer at work, where I have *real work* to do, and at home, in my *scant* free time. In both cases I don't have neither the time nor the inclination to waste days trying to conjure up a new workflow that is half as efficient as my current one just because some "designers" decided that they know better and that screwing up ten years of progress is a good idea.
And I stop here, because I could rant for hours about the groupthinking abort that GNOME 3 development process has become and I've already wasted too much of my free time.
Posted Feb 23, 2011 22:17 UTC (Wed)
by russell (guest, #10458)
[Link] (1 responses)
Posted Mar 17, 2011 17:00 UTC (Thu)
by dneary (guest, #55185)
[Link]
Dave.
Posted Feb 23, 2011 23:13 UTC (Wed)
by sorpigal (guest, #36106)
[Link] (11 responses)
I use my computer exactly how I want to use it and, incidentally, I have used exactly the same workflow for over a decade, keeping it across operating systems and window managers and desktop environments. It's totally irrelevant what the theoretical ideal workflow is, for me, because I evolved one that I like and know so well that no replacement is useful.
Any environment is acceptable insofar as I can configure it to match my way of working. This is the same reason I gave up on OS X: It did not and does not make it *possible* to adapt it to my workflow, not even as much as OS 9 did. GNOME has been on this worrying path for a while now, making changes that users aren't asking for without providing a trivial way to revert them because (I guess?) it is thought that users will just learn to live with it.
I use Linux partially because I am a tyrant. I want my way and only my way, no exceptions, and I throw a tantrum if I can't get it. I don't see why anyone else should have any say in how I use my computer, and I become quite emotional when someone suggests otherwise. I don't see why the aesthetics desired by a UI designer should lead to functional changes of any kind.
I don't want new. I don't want improved. I want control.
Posted Feb 23, 2011 23:21 UTC (Wed)
by aliguori (subscriber, #30636)
[Link] (10 responses)
As long as they have a "classic mode" toggle, I'll be happy.
Posted Feb 23, 2011 23:28 UTC (Wed)
by sorpigal (guest, #36106)
[Link]
Posted Feb 23, 2011 23:35 UTC (Wed)
by russell (guest, #10458)
[Link] (8 responses)
Posted Feb 24, 2011 0:20 UTC (Thu)
by njs (subscriber, #40338)
[Link]
Uh, this is FOSS. The people doing the evolving and the people doing the decreeing are the same.
Posted Feb 24, 2011 0:53 UTC (Thu)
by drag (guest, #31333)
[Link] (6 responses)
Who says they don't? It seems to me that Gnome folks program in the best manner they see fit and they design in the best manner they see fit.
> GNOME Shell should have been GNOME ShellS, where the S is a frame work for people to create shells how they like it (bring back configuration). Then just let the best design evolve out of that. But, no, they have decreed that this is the way everyone will work.
No. If you want to make your own 'shell' and let people use it Gnome isn't doing anything to stop you. It's all open source, modular, and you can do whatever you like with it.
Unity itself is a perfect example of this.
Your making it sound like if you don't use the default settings for Gnome then they will come out and shoot you in the head or something. It's really a wrong way to look at things.
Posted Feb 24, 2011 10:05 UTC (Thu)
by russell (guest, #10458)
[Link] (5 responses)
>No. If you want to make your own 'shell' and let people use it Gnome isn't >doing anything to stop you. It's all open source, modular, and you can do >whatever you like with it.
Typical, show us the code comment... Yeah I'll just knock one up in my lunch time to suit the job I've got to do in the afternoon. It will probably take my entire lunch because I have to start from scratch.
Really, I would like to just change a few configuration files instead. Maybe script some behaviour into the window manager, you know the stuff that you can do in a lunch time and share with your friends and colleges.
Eventually certain scripted behaviour, settings, themes, etc would become popular, highlighting what people really want. This works better than an academic process of predicting what people want.
Posted Feb 24, 2011 10:16 UTC (Thu)
by AlexHudson (guest, #41828)
[Link] (3 responses)
Posted Feb 24, 2011 12:22 UTC (Thu)
by sorpigal (guest, #36106)
[Link] (2 responses)
Posted Feb 24, 2011 13:11 UTC (Thu)
by AlexHudson (guest, #41828)
[Link] (1 responses)
Posted Feb 24, 2011 17:21 UTC (Thu)
by fmuellner (subscriber, #70150)
[Link]
Posted Feb 25, 2011 7:01 UTC (Fri)
by jmalcolm (subscriber, #8876)
[Link]
Unity is essentially a competing GNOME environment that, although providing a completely different user experience, shares the majority of it's ecosystem with GNOME 3.
XFCE is also a GTK+ based environment as is LXDE I believe. You can use these for the desktop experience if you like and still enjoy pretty much everything that GNOME has to offer and it will still feel pretty native.
In fact, I believe that one of the environments competing with GNOME Shell will be updated versions of Metacity and the GNOME panels. You can run the exact desktop you are used to now with GNOME 2 but with upgraded GTK+ and a lot more.
Beyond all that there are many different Window Managers and other environments. There is an embarrassment of riches when it comes to how you want to consume the parts of GNOME that you do find useful.
Demanding that we want to be able to configure and then wining that we would have to configure something to get back the behaviour we want seems silly. "I am a POWER USER dammit so you should design the system exactly as I like right out of the box!"
Pretending that GNOME is as restrictive and prescriptive as something like OS X or even Windows is misleading to my mind. I can run GNOME platform features and GNOME applications a lot of different ways without writing a line of code. I can also write a bunch of complimentary code pretty easily if I like. Thanks GNOME team.
I may or may not like the GNOME Shell. I will probably give it a try. I have not loved Unity based on the little I have tried it. I will probably run GNOME in a 'classic' configuration because, as with others here, that is what I am personally used to and feel productive with. What I will not do is to throw a tantrum because the GNOME developers feel the desktop can be done better and are trying to give it a shot.
After all, as far as I can tell, the GNOME team is explicitly providing the option to stay with the old metaphor if that is what you prefer.
I believe this is what the original poster was getting at.
Posted Feb 24, 2011 1:15 UTC (Thu)
by am (subscriber, #69042)
[Link] (7 responses)
How much this mirrors the sentiments of KDE 3.5 users a few years back is not even funny.
(And after trying to use KDE 4 for some years, I recently gave up and switched to XFCE :)
Posted Feb 24, 2011 5:48 UTC (Thu)
by AndreE (guest, #60148)
[Link]
Posted Feb 24, 2011 15:39 UTC (Thu)
by ThinkRob (guest, #64513)
[Link] (5 responses)
Sort of. KDE 4.0 got trashed for being a buggy, incomplete mess. That was true -- it was buggy and incomplete -- but that was kind of by design. The KDE devs stated numerous times that 4.0 was not meant to be a stable release for general use! (Of course why they chose to bump the major version for what was essentially a development milestone release is beyond me... but they did.)
GNOME 3, on the other hand, is getting trashed not for being buggy, but for adopting a variety of new UI approaches, killing off long-time features and flexibility left and right, all with (seemingly) little concern for what the core, long-time user base wants.
Now I'm not averse to trying new UI concepts, and I'm willing to bet a good bit of cash that anything that the GNOME designers can come up with is slicker than anything I've ever designed or will design... but I still don't think that change for the sake of change is a good thing. Who is asking for these new designs? What users are clamoring for a wild break from many of the conventions established over the last decade or so of computing? Why is this break so urgent, so crucial that it *must* immediately be implemented as the sole design of one of the largest DEs in the open source world?
</rant>
Adopting a whole slew of new UI styles and forcing a bunch of new UI principles on users all in a single release with no option to keep using the current design(s) seems to me to be tantamount to saying "we've decided this is better for you, so use it or leave".
My prediction is that many of the long-time GNOME users will do just that: leave. I expect XFCE and other DEs will see an influx of new users when GNOME 3 starts making its way into the major distros.
Posted Feb 24, 2011 20:37 UTC (Thu)
by mpr22 (subscriber, #60784)
[Link] (2 responses)
Because the changes were so drastic (I think some things had basically been rewritten from scratch) that it couldn't reasonably be called 3.anything. (Sadly, one can't get away with calling things "X.-1". ISAGN.)
Posted Feb 24, 2011 22:48 UTC (Thu)
by ThinkRob (guest, #64513)
[Link] (1 responses)
Fair point. That certainly makes sense. I do remember reading about how drastic the changes were.
Still, I can't help but wonder if maybe the versioning scheme that GNOME (and other projects) have used/are using wouldn't be better PR-wise (i.e. M.99 etc., where M is the current major version, becoming N.0 (N=M+1) when ready for general consumption).
Posted Feb 24, 2011 22:56 UTC (Thu)
by dlang (guest, #313)
[Link]
do you really want to have to explain to people how 3.4.5 is newer than 2.99?
Posted Feb 25, 2011 0:50 UTC (Fri)
by Frej (guest, #4165)
[Link] (1 responses)
Why are you writing all this without checking up on the facts? You might actually influence others who then could make a misinformed descision. What's your motive?
Posted Feb 25, 2011 23:12 UTC (Fri)
by GhePeU (subscriber, #56133)
[Link]
So the panels should be locked down to resemble the pseudo-panels of the shell, the applets shouldn't exist anymore (there were calls to completely remove them, even after they had been ported), etc. etc.
Just check the mailing list archives (desktop-devel and gnome-shell), this was discussed in January IIRC. Or https://bugzilla.gnome.org/show_bug.cgi?id=631553
I quote from the bug:
User: "Please don't diverge too much from how gnome-panel 2.32 works/looks. A lot of the people who will be using the gtk3 port are interested in it because it is a familiar setup, look and feel and not because of concerns over whether gnome-shell will work on their machines or not. In my setup for example, I have deleted one of the two panels (the top one)."
Designer: "Sorry, that is not what the fallback mode is designed for."
Posted Feb 24, 2011 11:58 UTC (Thu)
by Rehdon (guest, #45440)
[Link]
Rehdon
Posted Feb 24, 2011 19:24 UTC (Thu)
by debacle (subscriber, #7114)
[Link]
I really like GNOMEs idea of not having too many choices, buttons etc. since I'm working as a full-time system administrator for my mother (in parallel to my paid job, of course). We computer people don't have a clue how clueless people can misuse GUIs! Every removed choice or button is good.
However, I have my own preferences, which I can set in GNOME 2 via gconf, and I hope this choice is not taken away in GNOME 3. E.g. I like "focus follows pointer" and I hate "focus auto-raise", such things, that most people don't want.
Posted Feb 23, 2011 21:20 UTC (Wed)
by fb (guest, #53265)
[Link] (8 responses)
First he says: we know loads of people use minimization, and we should not remove it, causing pain to users, just to re-add it later because we didn't understand the workflow.
Then in the end, he seems to say: we don't fully understand the minimization workflow, we know many users need it badly, but what-do-i-care we are removing the buttons anyway, and if enough users are hurt by this, we'll re-add it later.
[...]
My two cents about this:
Oh, ops, terribly sorry. I got it now. This is a Linux desktop we are talking about. Only people with a 'computer related' degree deserves to use it.
Posted Feb 23, 2011 21:51 UTC (Wed)
by Frej (guest, #4165)
[Link]
I think it's going to hurt ;)
Posted Feb 23, 2011 23:59 UTC (Wed)
by otaylor (subscriber, #4190)
[Link]
It might seem a bit strange to make a big change based on only reasonable certainty, but more than that is hard to get. User studies can be informative, but in this area, we're really interested how experienced users work with a lot of windows, so the most basic approach of paying people off the street to sit in front of of a computer for an hour to do predesigned tasks wasn't going tell us much.
While shipping a half-hearted implementation of minimization might be the safe choice, it's not the choice that is going make progress. If we have half-hearted minimization, people who are used to minimization will continue minimizing, whether or not it's working as well for them as it did in the past. Given evidence that we have a reasonable plan, I was willing to trust our designers and move forward, rather than sticking with something just to avoid controversy.
Posted Feb 24, 2011 16:02 UTC (Thu)
by cruff (subscriber, #7201)
[Link] (2 responses)
Many people use both the minimize and maximize functions daily. I even run FVWM so that I can use the horizontal maximize and vertical maximize variants as well, with the vertical maximize winning in frequency over the horizontal.
I also heavily use the GNOME window list to manage the occasional gazillion terminal windows.
Posted Feb 24, 2011 19:12 UTC (Thu)
by halla (subscriber, #14185)
[Link] (1 responses)
Posted Mar 1, 2011 21:31 UTC (Tue)
by BenHutchings (subscriber, #37955)
[Link]
Posted Feb 28, 2011 11:04 UTC (Mon)
by nye (subscriber, #51576)
[Link] (2 responses)
Actually my experience is that a good proportion of people don't use minimize, and often don't understand the difference between minimize and close, because they lack the mental model that is required to grasp the concept.
Posted Mar 5, 2011 22:53 UTC (Sat)
by jzbiciak (guest, #5246)
[Link] (1 responses)
For some reason, your comment immediately reminded me of this. Is that really how we ought to be designing user interfaces for adults? Assume they lack the neuroplasticity to learn a new idea?
Posted Mar 22, 2011 16:46 UTC (Tue)
by nye (subscriber, #51576)
[Link]
Well, my last few years have been spent as a sysadmin, slowly eroding all hope and faith in humanity culminating in a total breakdown last year followed by the prescription of powerful antidepressants, so my perception is doubtless skewed, but my take-home observation is 'yes'. A significant minority - if not the majority - of adults indeed lack the neuroplasticity to learn a new idea, or to take an old idea and apply it in new contexts. They also lack the ability to reason, read, count, produce coherent sentences, and work cooperatively. Frankly, I'm amazed that some of the people I support can manage to feed themselves.
Posted Feb 23, 2011 21:45 UTC (Wed)
by Frej (guest, #4165)
[Link] (7 responses)
Maybe they could just fix 'maximize', since it's pretty useless for most windows to occupy the entire space. You often see windows (the os) with N windows (the abstraction) filling the entire workspace, just because maximize always does that. Has this changed in vista/7?
Suggestion:
Bascily, it should be an per program _autosize_ toggle, not maximize. I believe this is orthogonal to multiple workspaces, because it encourages a workflow where you can see more than one window at once.
Posted Feb 23, 2011 22:53 UTC (Wed)
by nix (subscriber, #2304)
[Link] (5 responses)
Posted Feb 24, 2011 0:33 UTC (Thu)
by Frej (guest, #4165)
[Link] (1 responses)
I guess the hard part is that this requires communication between WM/app instead of the WM just setting the new size of the window.
Posted Feb 24, 2011 8:19 UTC (Thu)
by nix (subscriber, #2304)
[Link]
I suppose this should really be an ability owned by a Sufficiently Smart Window Manager... but it isn't, it's part of the apps.
Posted Feb 24, 2011 0:51 UTC (Thu)
by sorpigal (guest, #36106)
[Link] (2 responses)
Posted Feb 24, 2011 7:08 UTC (Thu)
by obi (guest, #5784)
[Link]
Glad I'm on Gnome 90% of the time :)
Posted Feb 24, 2011 11:05 UTC (Thu)
by nix (subscriber, #2304)
[Link]
Posted Feb 24, 2011 10:29 UTC (Thu)
by MisterIO (guest, #36192)
[Link]
Posted Feb 23, 2011 21:52 UTC (Wed)
by gmaxwell (guest, #30048)
[Link] (1 responses)
Personally I use a tiling window manager (xmonad). I love it. It fits how I work. It's super fast and smooth, I can flip between programs, lay things side by side, and do this without losing my attention on the task at hand or even interrupting my typing. I could probably sustain 40wpm while switching between two windows every 10 seconds.
It's enough of an improvement that I'm willing to put up with all the incompatibilities it creates (and it creates a few the gimp sucks in this environment, things which attempt full screen or fixed size windows tend to go explody).
But my computer is basically unusable to anyone else. My SO, who is a competent GNU/Linux desktop user basically won't touch any of my systems. The multiple workspace and fixed position requires a good internalized model for how things are and can be organized. It's not something that is easy for people to get used to.
So, I'm now surprised to find that Gnome appear to be for once moving towards how I use my computer rather than rapidly against how I use my computer
and yet I still think they're making a bad call. I guess you can't make me happy.
Posted Feb 24, 2011 1:15 UTC (Thu)
by drag (guest, #31333)
[Link]
I have between 15-30 shells open, each one nicely labelled with the program running or the system it's on. I have a total of maybe 4 or 5 terminals open. I can flip between them with lightening speed and I rarely lose track of what is going on.
AND....
I can use Gimp just fine.
The secret is 'Gdevilspie' and 'tmux'. Plus a script called 'launch-gnome-terminals' that I created a *.desktop file for and stuck in ~/.config/autostart/ so that the environment gets rebuilt the same exact way every time I log in.
:)
I even use vim in server mode so that the files I am editing are always in the same place on the screen at the same time. And they recover automatically when I log out and log back in.
It's all quite fantastic and the environment is far more flexible and modular then people give it credit for.
For me a simple GUI environment is one that has a specific default configuration that aims to do 'the right thing' and relatively little. That way I only have to care about what I WANT to care about. Instead of having to figure out everything out I can get what I want through add-ons and plugins.
I used Gnome Shell for a little while and I found it to be quite a improvement in basic window and application management over the traditional WIMP-style interface. I don't see why I should be scared of it or whatever.
Performance was the main issue, but I am eager to try it again. I am very eager to get the 'whole ten yards' with Zeitgeist integration and such things. Also I am eager to see how it compares with Ubuntu's Unity.
Posted Feb 23, 2011 22:11 UTC (Wed)
by tjc (guest, #137)
[Link] (5 responses)
Owen's analysis of "Why do people minimize windows?" is incomplete. I minimize windows because I have my window manager set up to remove them from the alt-tab cycle when they are minimized, which makes alt-tab more useful. I use alt-tab because I am often forced to use OS X, and OS X has no convenient means to lower windows. (It has Expose, which, at least to me, doesn't qualify as convenient). So I have this alt-tab thing in my brain, and I drag it over to Linux, which is basically what anyone working in a multi-platform environment does; if not with alt-tab, then with something else. I could live without a minimize button if were to be replaced with a "depth gadget." In fact, I would be willing to trade both minimize and maximize for a depth button. The inability to control the stacking order of windows is the biggest shortcoming of most GUIs in current use. You can always lower a window that's in the way, but you can't raise a window that's entirely obscured. Alt-tab, cascading windows placement, Apple Expose--and to some extent minimize--are all attempts to work around this problem
Posted Feb 23, 2011 22:41 UTC (Wed)
by beagnach (guest, #32987)
[Link] (3 responses)
Posted Feb 23, 2011 23:03 UTC (Wed)
by sorpigal (guest, #36106)
[Link]
Additionally, in E16 the iconbox is optional. If you turn it off, too, then this means is that "iconified" is "completely invisible" until you open the desktop window list (which I find useful for drastically removing distractions).
Posted Feb 24, 2011 16:23 UTC (Thu)
by tjc (guest, #137)
[Link]
Sawfish
Posted Feb 25, 2011 13:30 UTC (Fri)
by dag- (guest, #30207)
[Link]
Posted Feb 23, 2011 23:26 UTC (Wed)
by sorpigal (guest, #36106)
[Link]
For the GIMP I'd like a way to make the floating palette windows appear under my mouse when I want them, then disappear when I don't. Alt+Tab switching them to in front of my image window(s) and then clicking the image window to "send the palette away" is my current best solution, but it's really not ideal. I've considered adding a "lower one" button to my title bars, but the problem of bringing them back remains. (You can have a LOT of GIMP palettes and image windows in your Alt+Tab window focus list.) I'd like a way to temporarily make the focused window be not there, like for 2 seconds, to give me time to raise a window that was behind it, then have it reappear stacked below the window I just raised.
When doing image manipulation it is a truism that you cannot have too much screen space, so keeping toolboxes always visible is just irrelevant. I'm sure this is true for other generally-full-screen activity. There has to be a better way to manage floating, overlapping windows when you have to have them all on one workspace. I don't think the Gnome Shell approach is a very good answer partially because it's very disruptive for a trivial window focus swap type operation. It seems to have been designed more for task switching than focus switching.
Posted Feb 23, 2011 22:40 UTC (Wed)
by lkundrak (subscriber, #43452)
[Link] (1 responses)
Posted Feb 23, 2011 23:18 UTC (Wed)
by otaylor (subscriber, #4190)
[Link]
Posted Feb 24, 2011 4:09 UTC (Thu)
by CuZnDragon (subscriber, #5097)
[Link]
And I am one of those that never use the maximize button. I have a large screen and have the settings so that just moving the mouse into a window activates it without raising it allows me to do a quick typing in a chat or command line without having to hide what I am currently working on. The only time I go full screen is with some games that are so dark they are hard to see in window mode.
I extensively use the minimize button for the way I work and I tried the workspaces and it was too much trouble when I needed something in the current space that I had open in another workspace.
Posted Feb 24, 2011 5:34 UTC (Thu)
by simosx (guest, #24338)
[Link]
Posted Feb 24, 2011 6:10 UTC (Thu)
by tonyblackwell (guest, #43641)
[Link]
I tend to have a lot of things opened and minimized, available at a single mouseclick. Perhaps not all at once, but often including my email, the browser while logged in to internet banking, my openoffice log of internet banking transactions into which I paste what I've been doing, a terminal for a myriad things, a terminal su'd, the next iso I'm downloading, the software update windows. Currently there is a nice index at bottom of screen as to what is what. Workspaces don't have that and icons within them are small and not all distinctive.
Maximize for movies, to get instant full-screen without having to muck around dragging it to the edge, still having other stuff distracting around the edge, having aspect ratio taken care of. Spreadsheets and file-browsing windows also benefit from maximize.
I like control of where I put stuff, and although not a Ubuntu user have shied away from the enforced dictates I hear of there.
Can gnome developers can be so wedded to an idea that they are comfortable just amputating tools I use every couple of minutes? I've the impression they ignore bug reports and feedback, so I guess this is only a little step down the road.
Posted Feb 24, 2011 8:04 UTC (Thu)
by Cato (guest, #7643)
[Link] (3 responses)
Not sure why this has happened with these projects - perhaps focusing on usability without enough weight given to compatibility? MacOS X and Windows haven't been through quite such radical revamps, probably because the product managers wouldn't let it happen.
Can anyone recommend a good alternative distro, ideally .deb based and a derivative of Ubuntu (which is out for me because of Unity in next release)? Maybe Linux Mint?
Posted Feb 24, 2011 8:09 UTC (Thu)
by lkundrak (subscriber, #43452)
[Link]
Posted Feb 24, 2011 8:54 UTC (Thu)
by patrick_g (subscriber, #44470)
[Link]
Posted Feb 24, 2011 15:37 UTC (Thu)
by hp (guest, #5220)
[Link]
also gnome 3 still has the old WM/panel as an option
Posted Feb 24, 2011 20:55 UTC (Thu)
by cyd (guest, #4153)
[Link] (4 responses)
I'm trying to keep an open mind, but removing the minimize button sounds like catastrophically bad judgment. How exactly is one supposed to temporarily dismiss a window? By shuffling it around to other parts of the desktop and/or another workspace? Yeah, that sounds a lot easier than clicking on a "minimize" button *rolls eyes*.
Posted Feb 24, 2011 22:46 UTC (Thu)
by dskoll (subscriber, #1630)
[Link]
I'm trying to keep an open mind, but removing the minimize button sounds like catastrophically bad judgment.
Indeed. I use minimize and maximize a lot. In fact, I use them so frequently I have them bound to keys (the exact same F-keys that the ancient SunView windowing system used in 1990... talk about imprinting!)
But I use XFCE, an environment designed to let you get things done rather than GNOME or KDE which seem designed to annoy the hell out of you while their creators fiddle about with dumb UI concepts...
Posted Feb 24, 2011 22:53 UTC (Thu)
by jthill (subscriber, #56558)
[Link] (2 responses)
Posted Feb 25, 2011 17:12 UTC (Fri)
by cyd (guest, #4153)
[Link] (1 responses)
Posted Feb 25, 2011 18:48 UTC (Fri)
by jthill (subscriber, #56558)
[Link]
More power to'em, it's a worthy effort, but the thing about limiting people's choices is you have to be right.
So I pretty much agree with you. The alt-space, "N" remark was just answering your question.
Posted Feb 27, 2011 7:49 UTC (Sun)
by blujay (guest, #39961)
[Link]
"I don't think it's generally a big deal to remove the maximize button." --one GNOME hacker
"I think having a second option 'Close Window' in the application menu if the application has multiple windows would solve this problem and allow us to get rid of **the visual clutter of a lone close icon in the titlebar**." --another GNOME hacker [emphasis mine]
I can't help but imagine these parallels:
"I don't think it's generally a big deal to remove the steering wheel." --GNOME hacker from 2111, when "the year of FOSS automobiles" is "approaching"
"I think having a second option 'Stop Car' in the automobile menu if the automobile has multiple speeds would solve this problem and allow us to get rid of **the visual clutter of a lone brake pedal in the floorboard**."
Now is when we "Rejoice for Choice," that we can let GNOME shoot itself in the foot and choose to move on to other software that gives its users power and control and freedom (like KDE and others).
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
AIUI the argument against "classic mode" is similar to the one against "beginner" and "expert" modes: they're bad for usability (confusing, etc) and make support and maintenance harder.
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
Great. Can you use javascript to put the minimize button back? Also, where is this documented?
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
(Of course why they chose to bump the major version for what was essentially a development milestone release is beyond me... but they did.)
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
1. how many _non IT-professionals_ do you know that uses workspaces in their desktop's workflow? (I don't know any)
2. how many _non IT-professionals_ do you know that uses minimize in their desktop's workflow? (most people I know do)
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
Instead have a toggle that switches between the last know position/manual resize, and a application aware optimal size, it's a bit harder when you have tabs, but it tends to work out anyway. At least metacity and maximize enlarges any browser to full desktop width, which is completely useless.
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
That's what MacOS does, right? I always wondered if there were any users who liked it.
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
Why do people minimize windows?
Why do people minimize windows?
Why do people minimize windows?
Why do people minimize windows?
Why do people minimize windows?
Why do people minimize windows?
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
I had tried a recent version by compiling from git and it worked quite well. So I look forward to the new version!
GNOME Shell 2.91.90 released
Minimize/maximize use description follows.
< Your option to page-down out of here! >
Alternatives?
Alternatives?
Alternatives?
Alternatives?
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
Window menu's still there, right? So alt-space, "N" should still work.
GNOME Shell 2.91.90 released
GNOME Shell 2.91.90 released
I actually quit on gnome because I got tired of trying to make it do what I want. The phrase that sticks in my mind from the linked email is "the GNOME 3 workflow". That says to me it's far from a joke: they're going down Apple's road.
GNOME Shell 2.91.90 released
Years from now...
--another GNOME hacker from the future, shipping out new driver UI theories without having done studies, testing, or thinking