[Ayatana-dev] new unity dep, libgnome-desktop-3-dev
Hi all, as of revision 1319 unity now depends on libgnome-desktop-3-dev, so be sure to have that until new packages are built that will bring that in. also, it depends on the latest libunity-misc that's all for now, -- -- Gordon Allott Canonical Ltd. 27 Floor, Millbank Tower London SW1P 4QP www.canonical.com ___ Mailing list: https://launchpad.net/~ayatana-dev Post to : ayatana-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana-dev More help : https://help.launchpad.net/ListHelp
Re: [Ayatana] Fwd: Place Shut Down as the last entry in the Sessions-Menu for Oneiric and beyond
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Jo-Erlend Schinstad wrote on 27/07/11 16:05: ... On 27 July 2011 13:39, Matthew Paul Thomas m...@canonical.com wrote: We have no evidence that a non-trivial proportion of people notice differently colored icons in the menu bar. You might say that's an assumption, based on personal experience and observation. Sure, it's not a scientific test, and I only have tested it on 10-20 people. But I noticed it the first time it changed color and I was surprised by how effective that really is. Yes, there seems to be a huge variation in the degree to which people notice small things on the screen. The sort of people who contribute to Ubuntu, for example, often find a colored icon in a panel quite enough to notify them of software updates. They can't understand how anyone would possibly not see it. But in user tests, nobody does. I don't see how they'd be more likely to search for them in the menu bar. * the indicators are visible. The launcher is not. * bluetooth is displayed as an indicator, not as a launcher. Fair points. * you often have many usb devices connected. * devices need a description. Menus as suitable for that. Launchers aren't. Those two don't seem to be related to the question of whether people will look to disconnect/eject a device in the first place. Personally, I would find more inviting a launcher that I could put anything into -- applications, bookmarks, documents, folders, contacts, whatever. I can do that with the Mac OS X Dock and (mostly) with the Windows 7 taskbar. Interesting. Wouldn't that make the launcher notifications (usb, etc) even more difficult to find? I don't know what you mean by launcher notifications. I thought I would want to keep all of my most used applications on the launcher and that I would temporarily move icons up and down according to what I'm doing so that I could rearrange keyboard shortcuts. Turns out I do, but in a much smaller degree than I had expected. The first six launchers are completely static. The next four is more dynamic, used for documents I'm working on right now, but don't need to launch using keyboard shortcuts. That works well for me, because then I launch and switch to my most used applications using only the left hand, and the shortcuts get very familiar. The work I'm doing right now changes, so I have to look up the shortcut number in any case, meaning that it doesn't bother me that I have to use two hands. In that regard, it would be nice if we could make a certain window type ungroupable, so that you could easily switch between open writer documents, for instance. ... Yes, there may be a useful distinction between document-based applications (for which it's best to show open documents in the launcher) and other types (for which it's best to show the application itself). - -- mpt -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4xKu0ACgkQ6PUxNfU6ecp7YQCfe8pC5/MqCOKG02IsuftUsYOy xecAnA+cjjGb83kOueieNcK+4Qsf6Tyf =DN65 -END PGP SIGNATURE- ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
Re: [Ayatana] Place Shut Down as the last entry in the Sessions-Menu for Oneiric and beyond
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Evan Huus wrote on 27/07/11 17:21: On Wed, Jul 27, 2011 at 7:39 AM, Matthew Paul Thomas ... This is what Havoc Pennington called the misguided 'hmm, maybe I can autogenerate my GUI' stage. http://www106.pair.com/rhp/free-software-ui.html Artificial intelligence researchers work on many things, but I'm not aware of any who are working on the problem of making an auto-generated settings interface anywhere near as understandable as a human-designed one. One project that is already sort of doing this is Wireshark. Individual protocols (TCP, IP, etc.) register their settings with very simple API calls along the lines of register_boolean(bool, Name, Description). Wireshark does the rest behind the scenes. The result is certainly usable, although probably not as usable as it could be. They're located under Edit-Preferences-Protocols in Wireshark if anyone actually wants to take a look. ... That's a great example of how terrible it would be if applied to every application. To be fair, there are several mistakes in the Wireshark presentation that could be fixed even while keeping it automated: checkbox labels are to the left when they should be to the right, numeric fields accept non-numeric characters, 1-of-many options are presented using radio menus even when there is plenty of room for radio buttons instead, the controls use tooltips when text-based controls never should, and so on. But there are many parts of that interface that just couldn't be designed well without a human. The ASN1 panel, for example, includes: * three fields for entering multiple port values, with no help in separating them * labels for those three fields which are mostly redundant with each other, because the presentation engine can't know how to (or even that it should) factor out the common words * a field for entering a file path, with no button for opening a file picker * a checkbox for Show full names, with no hint of what the alternative is. It actually reminds me a lot of WordPress 1.1, which also had an auto-generated settings interface. I designed a manual layout, and Matt Mullenweg implemented the redesign for version 1.2. It's far from the only reason WordPress then became wildly popular ... but it probably helped just a little bit. - -- mpt -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4xM54ACgkQ6PUxNfU6ecrICQCeLijkt0aZCExLNJz3a2er2keF 84kAn2A1IpeAIsHUVSfmzllFBLBuF8do =+O5u -END PGP SIGNATURE- ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
[Ayatana] New Alt-Tab
So I saw the video for Oneiric's Alt-Tab on OMG Ubuntu and I was a bit disappointed. It just seemed like icons with a bit of the coverflow tilting. It's probably too late to implement now, but I had the idea where Alt-Tab would make the launcher visible, non running applications would be desaturated and in the main view the window(s) for only the highlighted app would show (expose view if mulitple windows). This way, it stays consistent with look at the launcher for running applications and provides previews to the content of said applications. ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
Re: [Ayatana] New Alt-Tab
I love this. It really just makes alt-tab a keybinding for features we already have. Very clean, very simple, and enhances the idioms we've already started developing. ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
Re: [Ayatana] New Alt-Tab
Sure, nice idea. +1 2011/7/28 Alex Launi alex.la...@canonical.com I love this. It really just makes alt-tab a keybinding for features we already have. Very clean, very simple, and enhances the idioms we've already started developing. __**_ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/**ListHelphttps://help.launchpad.net/ListHelp ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
Re: [Ayatana] New Alt-Tab
Great Idea. On Thu, Jul 28, 2011 at 7:42 PM, Alex Launi alex.la...@canonical.com wrote: I love this. It really just makes alt-tab a keybinding for features we already have. Very clean, very simple, and enhances the idioms we've already started developing. ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp -- Owais Lone he...@owaislone.org http://www.owaislone.org ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
Re: [Ayatana] New Alt-Tab
It's a lovely idea, and three consecutive designers have dashed themselves on the rocks trying to get it right. I'd be thrilled if someone could do better! Here's what we found each time we tried it: *The launcher is spatial, the alt-tab is logical.* The alt-tab works best when it is a stack, with the most recently used stuff first. Toggling between apps is always a single alt-tab, and moving between a small group of apps scales up accordingly, alt-tab-tab-tab gets you to the third most recently used app/window. If you want to jump more than one step back in the stack, you want to be able to see where you are going. And this is the problem with the launcher, in order to give a sense of trajectory, you would need to reorder the items on the launcher. Which breaks people's sense of where things are and makes the launcher seem arbitrary. On the other hand, you could jump from item to item, but then you are not providing any clue as to where the next tab will send you. Which feels sucky (we tried it :-)). So, it's an interesting exercise and a very attractive idea, and if someone can make it work I would embrace the patch, but I think it's one of those attractive-but-wrong sinkholes. Prove me wrong :-) Mark On 28/07/11 15:52, Owas Lone wrote: Great Idea. On Thu, Jul 28, 2011 at 7:42 PM, Alex Launialex.la...@canonical.com wrote: I love this. It really just makes alt-tab a keybinding for features we already have. Very clean, very simple, and enhances the idioms we've already started developing. ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
Re: [Ayatana] New Alt-Tab
First of all, I'm no programmer. Second of all, I'm probably missing something. Third of all, how about making Toggles the Backlight Mode default? That way, it will be clear on first sight what alt-tab selections there are, especially if it only cycles through open programs. The white highlight used when keyboard focus is on the bar could be used (as well as showing the program window). An alternative would be to use the Scale plugin, in effect adding extra keyboard controls. Alt-Tab would scale all windows and cycle through them; to show which window is highlighted, use Zoom Window and Window Title from the Scale Addons plugin. This has the advantage of avoiding extra UI items or re-arranging the launcher bar while being somewhat intuitive. The problems would be how this would work on low-spec computers, and the work that would need to be done with Compiz (e.g. changing the behaviour of bindings, adding new functions, making Scale arrange windows in order of recent use c.). On Thu, 2011-07-28 at 18:04 +0100, Mark Shuttleworth wrote: It's a lovely idea, and three consecutive designers have dashed themselves on the rocks trying to get it right. I'd be thrilled if someone could do better! Here's what we found each time we tried it: The launcher is spatial, the alt-tab is logical. The alt-tab works best when it is a stack, with the most recently used stuff first. Toggling between apps is always a single alt-tab, and moving between a small group of apps scales up accordingly, alt-tab-tab-tab gets you to the third most recently used app/window. If you want to jump more than one step back in the stack, you want to be able to see where you are going. And this is the problem with the launcher, in order to give a sense of trajectory, you would need to reorder the items on the launcher. Which breaks people's sense of where things are and makes the launcher seem arbitrary. On the other hand, you could jump from item to item, but then you are not providing any clue as to where the next tab will send you. Which feels sucky (we tried it :-)). So, it's an interesting exercise and a very attractive idea, and if someone can make it work I would embrace the patch, but I think it's one of those attractive-but-wrong sinkholes. Prove me wrong :-) Mark On 28/07/11 15:52, Owas Lone wrote: Great Idea. On Thu, Jul 28, 2011 at 7:42 PM, Alex Launi alex.la...@canonical.com wrote: I love this. It really just makes alt-tab a keybinding for features we already have. Very clean, very simple, and enhances the idioms we've already started developing. ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
Re: [Ayatana] New Alt-Tab
On Thu, Jul 28, 2011 at 1:04 PM, Mark Shuttleworth mark.shuttlewo...@canonical.com wrote: It's a lovely idea, and three consecutive designers have dashed themselves on the rocks trying to get it right. I'd be thrilled if someone could do better! Here's what we found each time we tried it: The launcher is spatial, the alt-tab is logical. The alt-tab works best when it is a stack, with the most recently used stuff first. Toggling between apps is always a single alt-tab, and moving between a small group of apps scales up accordingly, alt-tab-tab-tab gets you to the third most recently used app/window. If you want to jump more than one step back in the stack, you want to be able to see where you are going. And this is the problem with the launcher, in order to give a sense of trajectory, you would need to reorder the items on the launcher. Which breaks people's sense of where things are and makes the launcher seem arbitrary. On the other hand, you could jump from item to item, but then you are not providing any clue as to where the next tab will send you. Which feels sucky (we tried it :-)). So, it's an interesting exercise and a very attractive idea, and if someone can make it work I would embrace the patch, but I think it's one of those attractive-but-wrong sinkholes. Prove me wrong :-) Mark This really surprised me: I've always found traditional alt-tab annoying for exactly that reason. While the logical representation of a stack makes a lot of sense in theory, I find I can never keep track of what's where on it beyond the absolute top anyways. Whenever I use traditional alt-tab for more than an immediate switch to the most-recently-used, I have to stop, look at the icons, figure out how many times to press tab, and then do it, which is slow. If I just press tab until I see the icon I want highlighted, I tend to miss it, and have to cycle round again to find it. Admittedly, I've never used the spatial alternative proposed here, but I imagine it would be much nicer if I could know 'four tabs is app X' all the time, rather than, 'four tabs is the fourth-most-recently-used, which was, uh, what again?' On the other hand, I do use a single alt-tab a lot, which would break if it was simply replaced by the spatial alternative. Maybe I'm unusual in this, or maybe a stack is still faster than spatial even though it feels slower. I dunno. I would be extremely interested if somebody did a usability and speed study on various window-switching methods though. Evan ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
Re: [Ayatana] New Alt-Tab
On Thu, 2011-07-28 at 17:20 -0400, Evan Huus wrote: On Thu, Jul 28, 2011 at 1:04 PM, Mark Shuttleworth mark.shuttlewo...@canonical.com wrote: It's a lovely idea, and three consecutive designers have dashed themselves on the rocks trying to get it right. I'd be thrilled if someone could do better! Here's what we found each time we tried it: The launcher is spatial, the alt-tab is logical. The alt-tab works best when it is a stack, with the most recently used stuff first. Toggling between apps is always a single alt-tab, and moving between a small group of apps scales up accordingly, alt-tab-tab-tab gets you to the third most recently used app/window. If you want to jump more than one step back in the stack, you want to be able to see where you are going. And this is the problem with the launcher, in order to give a sense of trajectory, you would need to reorder the items on the launcher. Which breaks people's sense of where things are and makes the launcher seem arbitrary. On the other hand, you could jump from item to item, but then you are not providing any clue as to where the next tab will send you. Which feels sucky (we tried it :-)). So, it's an interesting exercise and a very attractive idea, and if someone can make it work I would embrace the patch, but I think it's one of those attractive-but-wrong sinkholes. Prove me wrong :-) Mark This really surprised me: I've always found traditional alt-tab annoying for exactly that reason. While the logical representation of a stack makes a lot of sense in theory, I find I can never keep track of what's where on it beyond the absolute top anyways. Whenever I use traditional alt-tab for more than an immediate switch to the most-recently-used, I have to stop, look at the icons, figure out how many times to press tab, and then do it, which is slow. If I just press tab until I see the icon I want highlighted, I tend to miss it, and have to cycle round again to find it. I have to agree with Mark. In my last job (I am retired now) I had to make extensive use of the Alt-Tab feature. Often it was to find (search for) a browser window (organisation was stuck on IE 6), but where it really came into its own was when switching between two applications. My productivity would have been severely limited without that simple use of the Alt-Tab feature. Admittedly, I've never used the spatial alternative proposed here, but I imagine it would be much nicer if I could know 'four tabs is app X' all the time, rather than, 'four tabs is the fourth-most-recently-used, which was, uh, what again?' On the other hand, I do use a single alt-tab a lot, which would break if it was simply replaced by the spatial alternative. Maybe I'm unusual in this, or maybe a stack is still faster than spatial even though it feels slower. I dunno. I would be extremely interested if somebody did a usability and speed study on various window-switching methods though. Evan Tony ___ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
Re: [Ayatana-commits] [Merge] lp:~ted/libindicate/gtk3-and-stuff-like-that into lp:libindicate
small comments: - seems like you inverted the configure gtk2 and gtk3 checks? (i.e you check for gtk2 when building with gtk3) - shouldn't the indicate-gtk3-0.6.pc.in have -lindicate-gtk3 in its Libs rather than -lindicate-gtk? - Description: libindicate GTK stuff. ... can we get some better description? ;-) -- https://code.launchpad.net/~ted/libindicate/gtk3-and-stuff-like-that/+merge/69173 Your team ayatana-commits is subscribed to branch lp:libindicate. ___ Mailing list: https://launchpad.net/~ayatana-commits Post to : ayatana-commits@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana-commits More help : https://help.launchpad.net/ListHelp
[Ayatana-commits] [Merge] lp:~mterry/indicator-session/lightdm-updates into lp:indicator-session
Michael Terry has proposed merging lp:~mterry/indicator-session/lightdm-updates into lp:indicator-session. Requested reviews: Indicator Applet Developers (indicator-applet-developers) For more details, see: https://code.launchpad.net/~mterry/indicator-session/lightdm-updates/+merge/69686 LightDM's exposed org.freedesktop.DisplayManager interface seems to have changed. Here's the updates needed to make switching work. -- https://code.launchpad.net/~mterry/indicator-session/lightdm-updates/+merge/69686 Your team ayatana-commits is subscribed to branch lp:indicator-session. === modified file 'src/display-manager.xml' --- src/display-manager.xml 2011-07-05 06:10:58 + +++ src/display-manager.xml 2011-07-28 16:43:23 + @@ -1,17 +1,20 @@ !DOCTYPE node PUBLIC -//freedesktop//DTD D-BUS Object Introspection 1.0//EN http://www.freedesktop.org/standards/dbus/1.0/introspect.dtd; node - interface name=org.freedesktop.DisplayManager + interface name=org.freedesktop.DisplayManager.Seat !-- Show greeter to allow new login / switch users -- -method name=ShowGreeter/ +method name=SwitchToGreeter/ !-- Switch to a user, starting a new display if required -- method name=SwitchToUser arg name=username direction=in type=s/ + arg name=session direction=in type=s/ /method !-- Switch to the guest user -- -method name=SwitchToGuest/ +method name=SwitchToGuest + arg name=session direction=in type=s/ +/method /interface /node === modified file 'src/users-service-dbus.c' --- src/users-service-dbus.c 2011-07-11 10:58:39 + +++ src/users-service-dbus.c 2011-07-28 16:43:23 + @@ -195,15 +195,50 @@ create_display_manager_proxy (UsersServiceDbus *self) { UsersServiceDbusPrivate *priv = USERS_SERVICE_DBUS_GET_PRIVATE (self); + DBusGProxy *dm_proxy = NULL; + GError *error = NULL; + const gchar *cookie = NULL; + gchar *seat = NULL; + + cookie = g_getenv (XDG_SESSION_COOKIE); + if (cookie == NULL || cookie[0] == 0) +{ + g_warning (Failed to get DisplayManager proxy: XDG_SESSION_COOKIE undefined.); + return; +} + + dm_proxy = dbus_g_proxy_new_for_name (priv-system_bus, +org.freedesktop.DisplayManager, +/org/freedesktop/DisplayManager, +org.freedesktop.DisplayManager); + + if (!dm_proxy) +{ + g_warning (Failed to get DisplayManager proxy.); + return; +} + + /* Now request the proper seat */ + if (!dbus_g_proxy_call (dm_proxy, GetSeatForCookie, error, + G_TYPE_STRING, cookie, G_TYPE_INVALID, + DBUS_TYPE_G_OBJECT_PATH, seat, G_TYPE_INVALID)) +{ + g_warning (Failed to get DisplayManager seat proxy: %s, error-message); + g_object_unref (dm_proxy); + g_error_free (error); + return; +} + g_object_unref (dm_proxy); priv-display_manager_proxy = dbus_g_proxy_new_for_name (priv-system_bus, org.freedesktop.DisplayManager, - /org/freedesktop/DisplayManager, - org.freedesktop.DisplayManager); + seat, + org.freedesktop.DisplayManager.Seat); + g_free (seat); if (!priv-display_manager_proxy) { - g_warning (Failed to get DisplayManager proxy.); + g_warning (Failed to get DisplayManager seat proxy.); return; } } @@ -780,7 +815,7 @@ { g_return_val_if_fail(IS_USERS_SERVICE_DBUS(self), FALSE); UsersServiceDbusPrivate *priv = USERS_SERVICE_DBUS_GET_PRIVATE (self); - return org_freedesktop_DisplayManager_show_greeter(priv-display_manager_proxy, NULL); + return org_freedesktop_DisplayManager_Seat_switch_to_greeter(priv-display_manager_proxy, NULL); } /* Activates the guest account if it can. */ @@ -789,7 +824,7 @@ { g_return_val_if_fail(IS_USERS_SERVICE_DBUS(self), FALSE); UsersServiceDbusPrivate *priv = USERS_SERVICE_DBUS_GET_PRIVATE (self); - return org_freedesktop_DisplayManager_switch_to_guest(priv-display_manager_proxy, NULL); + return org_freedesktop_DisplayManager_Seat_switch_to_guest(priv-display_manager_proxy, , NULL); } /* Activates a specific user */ @@ -799,7 +834,7 @@ { g_return_val_if_fail(IS_USERS_SERVICE_DBUS(self), FALSE); UsersServiceDbusPrivate *priv = USERS_SERVICE_DBUS_GET_PRIVATE (self); - return org_freedesktop_DisplayManager_switch_to_user(priv-display_manager_proxy, user-user_name, NULL); + return org_freedesktop_DisplayManager_Seat_switch_to_user(priv-display_manager_proxy, user-user_name, , NULL); } gboolean ___ Mailing list: https://launchpad.net/~ayatana-commits Post to :
Re: [Ayatana-commits] [Merge] lp:~mterry/indicator-session/lightdm-updates into lp:indicator-session
looks good, thanks Michael -- https://code.launchpad.net/~mterry/indicator-session/lightdm-updates/+merge/69686 Your team ayatana-commits is subscribed to branch lp:indicator-session. ___ Mailing list: https://launchpad.net/~ayatana-commits Post to : ayatana-commits@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana-commits More help : https://help.launchpad.net/ListHelp
Re: [Ayatana-commits] [Merge] lp:~ted/libindicate/gtk3-and-stuff-like-that into lp:libindicate
Okay, updated. -- https://code.launchpad.net/~ted/libindicate/gtk3-and-stuff-like-that/+merge/69173 Your team ayatana-commits is subscribed to branch lp:libindicate. ___ Mailing list: https://launchpad.net/~ayatana-commits Post to : ayatana-commits@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana-commits More help : https://help.launchpad.net/ListHelp