Currently we are accessing the banshee database directly. This has the advantages of being very fast, and very flexible. It however is a bit evil. There are possibilities of crashing banshee, locking, etc. A better solution is a standalone daemon/dbus interface that reads the db, and that can shutdown when banshee starts up. This has a lot of other benefits such as using banshee's library code so that all our search align with the same searching that's already inside of banshee.

Many new laptop models come with two gfx chips, one with lower power consumption and other with more gfx processing power. What options exist to provide a better user experience to switch between the two, and what can be accomplished within the Precise timeframe.

Complete the Unity Greeter design

First session of the week, meet the team Review Oneiric

Once http://plus.google.com becomes more than just invite-only, I would love to see Gwibber integration (in the form of an installable .deb of a plugin) for it... Hopefully when the Google+ API comes out, it would be much easier...

With GNOME moving to using more and more systemd and PackageKit in different places, instead of having to patch all the apps using them to use our services, it would be great to implement those in our stack (aptdaemon and ubuntu-system-service). Work has been started to implement PK system bus interface in aptdaemon, so part of this is in progress already. Also, systemd interfaces needed so far are just a couple easy ones AFAIR, so shouldn't be hard at all.

We are seeing more and more people integrating on their laptop Unity and their preferred applications: - drag and drop support - custom quicklists - crazy integration patches It's proably time to get the most interesting one into the distro. Some effort were started to https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-o-default-apps-unity-integration, but due to lack of time, not anything was integrated in time. It's time to review the list and to push those forwards. If you have any idea on how to integrate better your preferred application with Unity, do not hesitate to suggest there! What is possible with current launcher is described in the Unity API page: https://wiki.ubuntu.com/Unity/LauncherAPI

Libreoffice online has been demoed by Michael Meeks at the Libreoffice conference 2011 in Paris. Lets discuss what this technology could mean for Ubuntu.

For Oneiric we went the easy route of adding a way to run external applications for the stuff we wanted in the control center. But this looks quite bad (some panels are embedded, others run a separate application), so for P I'd like us to do a cleanup and avoid as much as possible having external windows. So, here's a list of what we have: * jockey: this could perfectly fit in the System Info panel * language-selector: we just need ability to install languages and input method support in the region panel to completely remove this * software sources: also maybe as part of the System Info panel, which does updates? the tricky one is Ubuntu One, so I guess we'd have to leave it as it is, or write a panel that just embeds the u1-panel window?

* What can we expect from upstream this cycle? * Can we automate packaging more? * Worth using UDD branches? * Revise Ninja processes to be more efficient with macro upstream packages * Point releases as stable release updates?

It would be nice to improve the authentication mechanisms in Ubuntu to be more user friendly and make it easier to enable modern authentication schemes. This will probably involve: - Reviewing the messages/prompts in PAM for appropriateness - Adding hints to PAM to allow GUIs to better display the prompts (i.e. if a prompt is for a password, key number, if prompting for a password change). - Improving the Unity Greeter prompts to interpret the hints - Improving PolicyKit to interpret the hints - Making it easier to enable non-password authentication (e.g. LDAP, two factor).

Mike Sweet intends to remove the filters which are not used by Mac OS X from the CUPS upstream package: http://www.cups.org/str.php?L3930 We have agreed on this on the last OpenPrinting Summit and the filters will be continued and hosted by OpenPrinting. In addition, the filters for the PDF printing workflow will not be adopted by CUPS but joined with the CUPS filters we overtake. All Linux distributions would have to include this new CUPS filters package then to get CUPS continuing to work and have the same feature set as before. Due to the fact that this package will use the PDF workflow by default and that all major desktop applications send their print jobs already in PDF, this will complete the implementation of PDF as standard print job format in all distros (if they do not patch against the PDF workflow). See also my updated web page: https://www.linuxfoundation.org/collaborate/workgroups/openprinting/pdfasstandardprintjobformat This will also clean our heavily patched CUPS package somewhat. We need a name for this new upstream package. Should we call it simply "cups-filters"? Or "op-cups-filters"? Or do we rename Foomatic to openprinting and have the packages - openprinting-db - openprinting-db-nonfree - openprinting-db-engine - openprinting-rip - openprinting-cups-filters Note that we keep foomatic-rip/openprinting-rip separate from the CUPS filters as this filter is a universal filter which also works with many other printing systems. Another decision to make is whether we really should maintain all the filters which get over to us from CUPS or whether we should discontinue some. The filter set will contain "imageto..." and "textto..." filters which are not made use of by the usual desktop applications, they send all PDF (and some send PostScript). Alternatively, these filters could be made optional. Usually these are decisions to be made by upstream, in this case OpenPrinting, but in this session I want to have some pre-discussion from the distribution point of view, so that when this new upstream project gets created it will be created with the distro's needs in mind.

There's more work we can do to make Quickly better and keep pace with the platform.

How to increase theming consistency and how to properly style widgets and user interfaces

which ibus applet to use?

We would like to look into the possibility of asking people running -proposed to run a set of tests for each new SRU that will submit the results to launchpad. We would be doing most of the work, but I am not sure if the work required in the update manager is in your team or foundations

* Nepomuk by default? * KMail2/Akonadi * Akonadi off by default (calendar integration)? * QT_ACCESSIBILITY * LightDM?

We'll be presenting new features planned for 12.04 and hopefully we will be looking for support to implement the scrollbars on different toolkits.

How can we leverage and easier Unity testing from the community? How make them reporting high quality bugs? How can we make the barrier lower to contribute?

We should start creating Ubuntu API specification and documentation. At the moment if someone wants to write an application for Ubuntu, he has to search the web to know how to do common things. For example: how do I empty the Trash? How do I add my application to the notify area? How do I create a Control Panel window setting? More information on the wiki: https://wiki.ubuntu.com/UbuntuDevelopmentAPI

Printing out of desktop applications is done/managed by very many different dialogs, mostly depending on which GUI toolkit is used for an application. Some applications like OpenOffice.org have even their own dialogs. This is confusing the users a lot, having them to do the printing operation in many different ways. In addition, many dialogs are missing important features. Finally we got the project of implementing the Common Print Dialog funded and so it will be turned reality in Ubuntu Precise. See also: https://launchpad.net/common-print-dialog-gtk https://wiki.ubuntu.com/CommonPrintingDialog http://www.linuxfoundation.org/collaborate/workgroups/openprinting/commonprintingdialog

FreeRDP and Remmina to replace rdesktop, vinagre and tsclient

 * LTS?  * How to maintain quality * KDE papercuts? * starship troopers - whit?

Discuss and planify the One Hundred Scopes project : bringing *a lot* of scopes/lenses into USC for Precise.

Synchronising data from machine to machine and onto your own servers or cloud providers like Ubuntu One is a very useful idea. The Ubuntu One team have started work on u1db, a project codename for an easy-to-use database API layer which can work on any platform (Ubuntu, web servers, Windows, smartphone platforms) with the existing native databases (SQLite, MySQL, API layers, everything). Here we'll present the current ideas around u1db and how it will be used.

Having playlists shared between music players and between machines would be a great idea; make a playlist on Banshee on one Ubuntu machine and it also shows up in Rhythmbox, and Banshee on your netbook, and Ubuntu One music streaming. A standard format for playlists should be defined (m3u? pls? Something else?) and details of where those playlists should be stored, how they should be synced, and how music players can automatically read and write them seamlessly.

General session to discuss how and when new X.org bits will be integrated into Precise, and requirements from stakeholders. Starting with this LTS, we will be doing official X stack and/or driver backports going forward. We can work out some plans to coordinate this effort. Since this is an LTS, and since we can do post-release upgrades, we will likely want to be very conservative in selecting versions for specific packages. We anticipate spending most of our time during P on multi-monitor support and hybrid graphics (both covered separately), but if you have ideas for additional things to include in the X planning, please come and share your thoughts.

Thomas Senyk, Johanness Zellner, and Adam Weinrich from the Nokia Qt team will be at UDS to answer any questions and discuss topics such as the Qt roadmap. Thomas and Johannes come from the Professional Services area and are paricularly knowledgeable on on how Qt is used in embedded space.

Action items to plan a smooth integration of elementary apps with the unity experience

The Nokia Qt team would like to showcase how Qt is used in the embedded space. They'll be bringing a board running Qt on Ubuntu Core and talk about various open source and commercial/consumer products using Qt in non-desktop space. They will discuss how Qt roadmap is OpenGLES 2 centric with Qt3D, Qt5 Scenegraph, Shader effect enablement and other efforts. These efforts also have important implications for empowering the desktop.

In Oneiric we have introduced color management. We have at least basic functionality now. You can take existing ICC profiles and assign them to devices (AFAIK monitors, printers, scanners, webcams) and then the devices are appropriately corrected. The profile assignments are done via the Color capplet of GNOME Control Center, the management and distribution of the profiles via colord, applications use liblcms1 and liblcms2 to apply the ICC profiles. In Precise we need to complete and improve the infrastructure: ArgyllCMS needs to be made distro-ready and added so that color calibration devices get supported, libcms1 needs to be replaced by libcms2, as the former is buggy and not maintained any more, and a color management UI must get added to the Common Print Dialog. We also need to determine who will maintain the color management stack in Ubuntu.

In order to consider Qt a recommended choice for Ubuntu app development, we need: - Better integration of the Qt development tools with Ubuntu-specific processes and tools, such as packaging, bzr, Launchpad... - Ability to access the Ubuntu platform APIs from Qt. - Ensure the packaged Qt development tools are in shape We'd like to discuss what needs to be done and a roadmap for this work.

Allowing people to sign in to an Ubuntu installation with their Ubuntu One credentials has a number of benefits: it makes sense on smaller or attached devices, your device is automatically paired with Ubuntu One, meaning that you have all your data without an additional "pairing" step, it means that you don't have to remember a different password per machine (not necessarily a benefit in everyone's eyes). This session is to discuss how, technically, signing in to an Ubuntu installation with Ubuntu One credentials might be done, what it would mean (does a new "local" user account get created? Should the "local" user account already exist and be matched somehow with the U1 account?), and which issues are generated by it (local caching of credentials, logging in while offline, etc).

This is where we will track the progress of Kubuntu Docs for the 12.04 LTS release of Precise Pangolin.

There are different services to print through the cloud, which means that you either print out of a cloud application, like Ubuntu One or Google Docs, or from a mobile device through the cloud to a printer at home, in the office, or at a print service. We should support this in Ubuntu One, allowing to print printable files from the cloud storage to services like Google Cloud Print and/or HP ePrint.

Ensure the ability to fully use Ubuntu Medium in Ubuntu 12.04 This is needed for the typographic review of the user interface. It will be necessary to be able to ship and use Ubuntu Medium and Ubuntu Medium Italic without advise effects on the rest of the interface. Currently a metadata loss occurs in FontConfig, the font matching framework used on Linux. Data that is (incorrectly) presumed to be duplicated is discarded, later causing the substitution of Ubuntu Medium with Ubuntu Bold or vice-versa at higher up levels of the stack in Qt and Gtk+. Similar issues occur with the Linux Libertine typeface, another advanced font with multiple weights within the same family. To implement, this will require upstream fixing of FontConfig and associated libraries to fix the model. To encourage completion or this, Ubuntu Medium will be once again re-uploaded to the 12.04 archives at an early stage.   LP bug #744812. There's a data-loss (discarding additional meta-data on the assumption that it's duplication, when it's not) in FontConfig, which is perpetuated by anything that follows FC's data-model and presumptions (Qt, Gtk+).   I keep not getting far enough stuck to work out a decent solution. But two guesses at the moment are either to duplicate entries (give two slightly different entries) within the current model, or to extend the data-model to cope with the meta-data in the fonts themselves.   I believe that other OSs have two basic API entries; a "dumb" mode that just matches on string:name + bool:bold + bool:italic, and an advanced selection mode that selects based on string:name + integer:weight + integer:slant + string:style + etc.   The two advanced open fonts it affects are Ubuntu (shows up particular with Medium) and with Linux Libertine.

Google Cloud Print allows printing on your local printers from whereever you can send jobs to the Google Cloud Print service, especially from mobile devices, like smartphones, tablets, and computers running Google Chrome OS. Google itself only offers to either use HP ePrint printers (network printers which can receive print jobs by e-mail) or to use the Chromium browser as server to share the local printers to Cloud Print. The latter is officially only supported under Windows and Mac OS X, but under Linux it can be enabled through a secret "Cheat Mode", entering the URL "about:Flags" and enabling the “Cloud Print Proxy” entry, then clicking the wrench at the upper right and selecting "Preferences" -> "Under The Hood" and there will be a "Google Cloud Print" section near the end of the list. So these methods are awkward, running a full-featured browser only to make available ones printers or to have to use a printer out of small choice of models. There is also a "cloudprint" package in Universe which is a small lightweight command line tool to replace the Cloud Print functionality of Chromium, but usual desktop users are not aware of that. What we would need is something like a capplet in GNOME Control Center to (de)activate and configure Google Cloud Print. This GUI needs to be designed and it also needs to be decided which user daemon will be used to enable Cloud Print, the one of Chromium or the "cloudprint" package.

Community testing of accessibility during the development cycle. During most cycles, testing can not be started until the last month or later of the cycle. Testing needs to start as close to the cycle start as possible, with Alpha2 as the latest date to be able to test it.

With our switch to tb this cycle we let the default installation without a calendering solution. We should decide if we need one by default. We might not need a full client ui but we should probably at least have a way to subscribe to a remote calendar to have desktop integration and notifications working

Using 1.5GB DVD Image by default

kde filesharing improvements/fixes

LTS Thunderbird Enhancements

discuss what GNOME version we will track for the LTS.

For some derivative products based on Ubuntu, it is key to be able to offer some customization of Unity. For example, business and government entities running Ubuntu would prefer to have other applications on the home lens than photo viewers and music players. It is also important for localized images for regions like China.

* Whither Kubuntu Mobile? * Plasma Active for the win?

General improvement of power consumption

We currently have a couple of problems with the Firefox and Thunderbird upgrade experience, which users of Mozilla's update service don't experience (ie, everybody on Windows, Mac, or anyone using mozilla.org binaries on Linux). 1) A long standing issue we have is that upgrades totally break all currently running instances of Firefox or Thunderbird until they are restarted. This is made worse because we move the install location around between updates, but pulling the rug from underneath any running instance is probably never going to work reliably, even if the install location doesn't change. Note, this isn't just an issue with Mozilla applications - we just notice it more because we update them a lot more frequently than anything else in the archive. As another example, there was a glade -> GtkBuilder transition a few cycles ago, where upgrades between distro-versions broke things such as the currently running instance of gnome-panel [2] due to the glade files being replaced with gtkbuilder ui files. I just wanted to point this out before people start blaming Firefox that this is really a problem with how our package manager works... Upstream are currently planning work around silent updates, and they have a wiki page documenting how their update process will probably work [3]. We might be able to learn some things from this. 2) We have no way of warning people if any of their addons might stop working when we offer them an upgrade. People using upstream builds are notified before an upgrade if any of their ddons aren't compatible with the update, and if there are no addon updates available to fix this. Do we want to provide such functionality, bearing in mind: - We haven't had that many bug reports yet from users complaining of addons being incompatible after an upgrade, although there have been 1 or 2. - As of the time of writing, 92% of addons on AMO are compatible with the current Firefox beta (due to be released on November 8th). However, approximately 75% of addons in use are not hosted on AMO [4]. - From Firefox 10 (Jan 31st, 2012), addons are going to default to being compatible, with some exceptions (addons with binary components) [4]. [1] - https://blueprints.launchpad.net/ubuntu/+spec/desktop-p-mozilla-upgrade-experience [2] - https://launchpad.net/bugs/422568 [3] - https://wiki.mozilla.org/Background_Updates [4] - https://wiki.mozilla.org/Features/Add-ons/Add-ons_Default_to_Compatible

CouchDB replication through U1 makes Ubuntu a unique and attractive platform for forward-looking apps. However, reliability problems have made it difficult to actually use. So how do we improve the situation? And what can app devs expect from CouchDB sync in Precise? * What current upstream CouchDB work might improve the situation for P? * What version of CouchDB will ship in P, what will the U1 infrastructure run? * Although an understandable move because of load/reliability issues, syncing only once an hour makes the platform far less interesting from an app developer perspective... are we stuck with this? * From a user experience perspective, what we really want is continuous replication... and that's also where the platform gets *really* interesting [insert awesome Novacut demos]. * Jason would like to see a U1 sync testbed available in parallel to the production infrastructure... uses usual SSO, but runs on newest Ubuntu release and newest CouchDB... has *no* data-safety guarantee! I want to help improved U1 CouchDB reliability, but can't do much if the production infrastructure isn't usable for my apps.

Text-Free Boot

That's another "classic" for UDS, we should probably not change a lot but we said we would revisit having pitivi on the CD, we should also have a look to GNOME new stuff like the sushi previewer for nautilus

General discussion session about all topics related to LibreOffice packaging and goals for Precise.

How will we build the list of desktop bugs that we think we should work on the during the LTS cycle? The current list from Pedro and Qa are a nice but they only cover selected set, how do we get visibility on i.e foundation issues, kernel bugs, plymouth issues, etc that impact on what we are doing?

With 11.10 we released a localized image for China. This image contained many customizations to make Ubuntu more appealing to users in China. This included significantly improving support in CJK languages, adding gwibber plugins for services popular in China etc. There are still some improvements that can be done to cary this work forward for our Chinese community.

From Oneiric on we are using GNOME 3.x and from this GNOME version on there is a printer setup tool in the Control Center. This tool is new and has much less features than system-config-printer and especially does not contain all the functionality for selecting best drivers, identifying network printers, HPLIP integration and so on. So dropping system-config-printer in favor of GNOME's new tool will be a major step back. Discussion has already started on the OpenPrinting Summit in April this year (see link below) and on the UDS for Oneiric. Here we will continue discussing the situation. https://www.linuxfoundation.org/sites/main/files/system-config-printer-status.pdf In Oneiric we have continued to use system-config-printer, patching and configuring gnome-control-center and the system menu so that system-config-printer is started in its own window instead of the capplet of the control center, at least for Unity desktops. For Precise we want to use the original GNOME tool but an improved version. For this Tim Waugh, upstream author of system-config-printer has already put all algorithmic parts of system-config-printer (identifying CUPS' printer auto-detection results and finding out which are from the same physical printer, finding best driver, prioritizing drivers, HPLIP integrations, ...) into a D-Bus service which any other printer setup tool, independent of the programming language in which it is written, can use. This way the GNOME tool can be improved to have the needed capabilities to make printer setup easy for everyone. No the printer setup tool needs to be modified to make use of this new D-Bus service. Depending on how far the tool gets developed, we will make it the standard tool in Precise and this way have one standard printer setup tool with perfect integration in the System Settings in both Unity and gnome-shell environments. See also: https://bugzilla.gnome.org/buglist.cgi?component=Printers https://bugzilla.gnome.org/show_bug.cgi?id=654742

Fregl tells us how to make it not crash

All the open issues that need to be fixed lo-menubar integration into main.

Since Unity has been introduced, many strides have been made with accessibility. Unity 3D is approximately 80% accessible, so far as official oneiric packages go, and unity-2d is 99% accessible, thanks to the great work from the QT a11y folks, and the unity-2d developers. Both are likely to be 100% accessible for the LTS, which is great. However, they are not the only areas of accessibility that need improving. I have already started making a work item list of issues to address, to give the desktop more pollish when it comes to accessibility. Such issues include: * Making sure all indicators have accessible descriptions/labels. * Making sure any icons in indicator menus that need accessible descriptions/labels have them, a prime use case is network manager, to indicate wireless network strength. * Further pollish on Ubiquity's accessibility, particularly selecting a timezone. * Connecting up the rest of the accessibility profiles in the ubiquity a11y indicator. * Further pollish of the unity greeter accessibility experience. * Make sure the accessibility experience between unity 2d and 3d is the same. The list I have is slightly more technical and goes into more detail, but what is listed above is the gist of what I have. Whether this needs a session at UDS I don't know, but a blueprint will likely be needed for work items tracking.

It would be nice to be able to optionally run Wayland in 12.04 to try out the technology. This will involve: - Modifying LightDM to support Wayland - Writing a Wayland compositor - Running an X server that writes to the compositor - Making it easy to enable this - Disclaiming all responsibility This would be a low priority task.

Include OpenSC in Ubuntu for smartcard authentication

  * Improvements to Muon?   * Samba install?   * Amarok etc codec install   * Distro Upgrade? 11.10 upgrade notification and 12.04 to 14.04 LTS notification upgrade

For two release cycles, gnome-session has deliberately not done session restoring on login. This is a huge stumbling block for power users, who have to spend a lot of time on each login re-setting up their work environment. This should be solved for 12.04.

- using the vala rewrite for p? - no full text search - datasource to consider? - privacy concerns - gnome-activity-journal in control center.

On release day, users are offered to upgrade their computer to the latest release. However, there are a number of ways in which upgrades don't complete successfully. What can we do to make upgrades safer, more reliable and more pleasant? See bug http://pad.lv/876146

In the real world there are always going to be failures, triggered by things like software bugs, hardware failures and misconfiguration. Ubuntu should where possible handle common failures and provide predictable feedback to the user that the system is broken.

Making our multi-monitor, dock and projector support better.

There are a few things we should cover for NetworkManager this cycle: mostly polish, but as general ideas: - reinstate daily builds for 0.9 to help with testing, knowing in advance about bugs, etc. - testing some of the new features likely to land and be useful for Precise: bonding, VLAN support, etc. - types of bugs or small enhancements that should be looked at in priority ?

For USB-based mobile broadband modems it got common that the Windows driver is stored on the modem itself and as long as the driver is not installed the modem shows itself to the computer as a USB mass storage device. Therefore usb-modeswitch got introduced to switch these modems into modem modeunder Linux as there the driver is already available as a kernel module shipped with the OS. A similar problem occured now for USB printers, they also show themselves as mass storage devices by default to provide their Windows driver without need of a CD. So usb-modeswitch needs to be extended to printers Tests should be done whether the switching mechanisms are the same as for modems and only the printer's IDs need to be added or whether more work is needed.

Improve finger print authentication in Ubuntu

On that topic something which comes often is when to lock screen, we should do a table listing the different cases where the screen might get locked (i.e on idle delay, suspend, lid close, user switching, etc) and the different credential scenarios (i.e autologin, password login, etc). We should also decide if the "lock screen on idle" option from the control center should affect, i.e user switching The topic comes quite often in bugs report and there is some conflicts there being "security" and "no password prompting", we currently force the screen locking on user switching or suspend but we should probably give an easy way to opt out for users who don't want it.

having a lot of thoughts on what can be OneConf (even it the scope of the post is more than a cycle I guess), I gathered them on this blog posts: http://blog.didrocks.fr/post/OneConf-in-Oneiric-and-the-way-forward%E2%80%A6

treach Darkwing how to package

Since we have been a bit sloppy on that topic since lucid (or busy with new features) we should aim back at better performances for the lts, that include cpu and memory usage and login time. See https://bugs.launchpad.net/ubuntu-boot-speed https://wiki.ubuntu.com/DesktopTeam/11.10/BootSpeedAnalysis has boot charts for various releases and comparative charts for lucid vs. oneiric and Dell Mini 10 (reference platform, slow Atom + SSD) vs. ThinkPad X201 (fast quad core + SSD, for identifying unnecessary dependencies) https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0ApM3WkUQgY4WdGtnMFFnelU2ZlpMSnUxMmhFZGhWZ0E&authkey=CL-LzIQL&hl=en_US#gid=0 has more comparisons of two machines between 9.10 and 11.10.

Say cheese