Contributor HDave

F-Spot is a relatively clumsy photo-manager. One major issue is the fact that it completely ignores my organization of photos into folders and imposes its own schemes and tries to make a duplicate copy of all photos to "import" it into the F-spot collection. Why can't the organizer respect the organization that the user already has and build up on that?

gThumb is so much better in this aspect. Moreover, you can also choose to work with your collection in Nautilus if you feel like, without having to open the photo manager every time you need to touch your photos.

It would be nice to have an "Eject Device" Button in Nautilus, in the top toolbar, by:

[Back] [Forward] [Up] [Stop] [Reload] | [Home] [Computer] | [Search] [TimeVault] | [Eject Device]

That way, I wouldn't have to navigate to the Desktop to eject a device. In order to avoid confusion, I would hide the button if the user is not browsing a mounted device.

Normally Linux (and other operating systems) have support for only 8-bit color per channel, resulting in a color depth of 24 bits.

Add support for 10, 12 and 16 bit per color channel, so we can have a color depth of 30, 36 and 48 bits.

This will allow for HDR (High dynamic range), medical imaging, etc. This will make Ubuntu great for professional use, graphics, science, medicine, etc.

The ATI FireGL V7350 graphics card supports 40-bit and 64-bit color.

Instead of just 16,777,216 colors, we could have over 4,294,967,296 distinct colors!

http://en.wikipedia.org/wiki/Color_depth

Steal those ideas (and open source code)*.

Look for benefits in the usability + administration + performance + security features and enhancements from other Linux distributions and incorporate them into Ubuntu.

CentOS5.1, RHEL5.1, SLES11, and TurboLinux beta benefit from substantial investment in server features and including management, virtualization, drivers, throughput tuning, security, SELinux configuration, etc.

Fedora, OpenSUSE, Mandrivia, Linspire, etc. include nice enhancements in desktop performance, configuration, layout, themes, and administration.

For some reason there is a great deal of re-inventing the wheel in these areas where the "upstream" development of a management tool, feature, or performance enhancement is "owned" by a "competing" distribution *even though those tools are open source*!

It seems as if every distribution duplicates effort for getting their systems to boot faster, be more robust, tweaking Xorg, laying out and naming menu items, etc..

Perhaps a wiki could be dedicated to capturing these features for potential incorporation into Ubuntu?

* Technically it isn't stealing since we are talking about OSS with compatible licenses.

It would be a huge step for linux if the use of MS office in wine would be easier. There are tricks now that can be used to make it work, but the average PC user will not be able to pull it off. If it is possible with commercial emulation software, why shouldn't it be made easier with Wine.

You can wine that we must all use Open Office. However, my very own brother refuses to use ubuntu solely because he can't use msoffice.

There are many misconceptions about Linux and the open source operating systems that are preventing many people from adopting this technologies. A way to solve this could be launching some kind of global Linux campaign in order to emphasize the benefits of the Linux operating systems. For instance:

- Free. Because we believe operating systems should be free. No licensing fees
- Free Upgrades. New features are free
- Friendly community
- Developers who listen. If you need something, it is more likely you will be able to get the feature added
- Large amount of high quality, free software
- Find the linux OS for you, each has different advantages, optimised towards different users.
- More customisable. Find the desktop for you
- Runs Openoffice & firefox
- Manages your software for you. You no longer need to worry about checking for updates yourself. Linux does it automatically for you, and lets u choose what to upgrade
- Developer friendly

More than specifying a lot of benefits, the idea would be concentrating in the most beneficial aspects, while fighting the most widespread wrong ideas about Linux.

As this campaign would benefit all the parties using in Linux operating systems, it could be funded by a group of them. This would help reduce the economic cost of this campaign to Canonical/Ubuntu. Moreover, as one of the most common misconceptions about Linux is that there is an excessive fragmentation, this would help giving a unified and strong view of Linux.

Many thanks to Auzy for the reformulation of the benefits!

Ways to make terminal server (remote desktop) client better:

A) make it possible to switch out of fullscreen mode without disconnecting

B) add a maximize option to use up as much of the screen as possible (without going to full screen mode)

C) allow custom screen resolutions in the display setup tab

D) add scroll bars or some way of scrolling if the terminal client screen size is smaller that the size of the host desktop or make the client report its screen size so the host desktop can adjust itself

Fantastic article on consistent user interfaces. This really should be a blueprint on how to design the next release.

http://www.smashingmagazine.com/2008/08/12/top-10-usability-highs-of-the-mac-os

Please, add a gui front-end to hwinfo to ubuntu.
The only similar utility I have found in Ubuntu repos is sysinfo but it's a minimal gui and not very usefull.
I think that a gui that shows every aspect of system hardware (like the info printed by hwinfo) is a mandatory tool for every linux distro.

gnome-system-monitor provides a cleaner interface than Windows "Task Manager", and provides many useful features including list of processes running, memory and network usage, etc.

This functionality should be more accessible by key binding it by default to CTRL-ALT-DELETE as Windows does--as this is more familiar to users coming from Windows. The current key binding for CTRL-ALT-DELETE brings up the shutdown/logout options, which is already accessible as a desktop button, which is redundant for a relatively less used function.

gnome-system-monitor is an idle interface for monitoring the system and killing processes without using the terminal and provides an interface for doing this that most are already familiar with. Changing it's key binding would make it's functionality much more accessible.

Currently when a program falls into an infinite loop and hangs, grabbing nearly all the cpu cycles when it does, what's the current procedure for the end user? The computer will most certainly be painfully slow to user input (assuming it's responsive at all.) We need to kill the offending process ASAP. This can be daunting for new users, and frustrating even for seasoned users. Currently the options include xkill, killing from the System Monitor, ps then kill, or (one of my current favorites) pkill. Sometimes X is so slowed down, I even Clt-Alt-F1 into a fresh tty. In the 21st century can't we do better?

What about this instead: When a runnaway process grabs nearly all the CPU cycles, 3 seconds later a 'ballon' notification should appear in the taskbar informing you that the system suspects this, and offering you the option to 'click here to kill the program'. Offer this functionality as an option that can be turned of if the user so choses.

Why not have the system realize what's going on, 'quaranteen' the damage by automatically renicing processes, and provide a notification with option to terminate the offending process?

In the new way I have suggested, the system monitors it's processes. If a process belonging to a currently logged-in user suddenly grabs 95% (just for example) or more CPU and keeps hold of it for more than, say 3 seconds, a few things happen. First, the offending process is automaticallly re-niced to the highest nice value. Second, a freedesktop.org - compliant singal is sent. Third, GNOME, KDE, XFCE, or whatever window manager / desktop enviro is running picks up the signal and creates an alert to the user. The processes for sending the signal, receiving it and displaying the alert have all been re-niced to the lowest nice value, effectively clearing the way.

Maybe the alert will take the form of a 'balloon' from the taskbar? Maybe a nice compiz fade-in popup? Maybe something like mumbles-project.org uses? Whatever method is used, the alert notification must supply a 'one click' to terminate the offending process.

What we're doing here is changing the paradigm. The question isn't 'How do we allow users to terminate a misbehaving program?' Intead its 'How do we ensure users's systems will remain responsive to input.'

[....]

It has become obvious that Grub2 needs some extra programming help, as Grub-legacy will no longer be adding new features and the end of Grub2 is nowhere in sight. Canonical should allocate 1 or 2 programmers for a few months to help complete grub2 quickly, so that X86 support might be complete in time for Interpid Ipex.

One could argue that the boot loader is the most important part of the operating system, because if it doesn't work properly (which it hasn't been for some of us), it can prevent every OS from working on the computer. We should treat it with respect, and help them. Despite grubs importance, programmers generally enjoy working on more exciting projects like Gnome or KDE, which is one reason why development is slow.

Grub2 fixes a lot of previous brainstorm ideas including many booting issues (like mine), so completion would close a lot of bugs, whilst also making Ubuntu more user friendly because Grub2 deals with booting issues better.

Voting for this ensures that Canonical allocates some developers to the Grub2 project, which would be a big win for both Ubuntu and linux in general!

Planned Features for Grub 2
* Rescue mode saves unbootable cases. Stage 1.5 was eliminated.
* Dynamic loading of modules in order to extend itself at the run time rather than at the build time.
* Graphical interface.
* Fix design mistakes in GRUB Legacy, which could not be solved for backward-compatibility, such as the way of numbering partitions.
* Scripting support, such as conditionals, loops, variables and functions.
* Cross-platform installation which allows for installing GRUB from a different architecture.
* Internationalization. This includes support for non-ASCII character code, message catalogs like gettext, fonts, graphics console, and so on.
* Portability for various architectures.
* Modular, hierarchical, object-oriented framework for file systems, files, devices, drives, terminals, commands, partition tables and OS loaders.

[....]

cdroms are slow, don't allow to modify their content easily, they are weak and not as easy to carry as USB keys.

It would be great to provide Ubuntu as a liveUSB just as Mandriva does. We should be able to carry our distribution on any computer, manage our preferred settings (do I want binary drivers enabled ? what is my preferred resolution ?), etc.

The thing is not to make something transportable, but really a nomad system that could be used just the best way as it could on any computer.

There should be a single, low cpu app that displays all the notifications for all programs in a corner of the screen (bottom right)

It should say:

[App Title]
-----------
[Notification]

It is much better than popups and notification Icons.

Eyecandy would also be nice here - fade effect, etc.

Dansguardian is a web content filtering tool to prevent children from accessing unsuitable websites. It is not 100% perfect but indeed a VERY GOOD tool, to protect children.
www.dansguardian.org

I use dansguardian for a while now.
I would appreciate a better integration of dansguardian within ubuntu (like in ubuntu CE).
I also like the GUI of dansguardian which was extra designed for ubuntu CE.
I think ubuntu should also be for little(!) human beeings, therefore a tool like dansguardian is very important to spread ubuntu even more.
If not integrated within ubuntu it should be integrated in edubuntu (and therefore available for all ubuntus via packet-management).

Remove Mono and dependent applications from default Ubuntu Desktop CD. Mono occupies a significant amount of the valuable space on the live cd that could be used for translations and other things. Applications using mono use much more memory than their non-mono counterparts. Functionality can be provided by other applications that are just as good.

This will NOT remove Mono or any of the applications from the Ubuntu repositories, just the default Desktop CD. (Although removing them from the CD may mean they don't need to be in Main anymore)

This affects two applications included by default: tomboy and f-spot. Tomboy can be replaced by either sticky notes or zim and f-spot by gthumb.

I've got two monitors, and right now if I use the wallpaper settings in gnome it stretches the wallpaper across the two monitors. It does this even for the default ubuntu wallpaper. This can look really bad depending on the image. In order to get around this I had to get two wallpaper images and join them into one large one with the GIMP and set that as my wallpaper.

There should be a way in gnome to set a different wallpaper for each monitor.

==== Merged with ideas of a similar scope: Different wallpapers on different workspace ===

Many workspaces option in Linux allows you to separate your work, why not allow users to have different wallpapers on each workspace. This allows for easier identification of which workspace you are on.
When used with compiz-fusion you can rapidly switch between your workspaces and the different wallapaper will allow you to quickly recognise what desktop you are on.

Currently this is natively available in KDE but not in Gnome. In Gnome you can't even let Compiz take over the wallpaper control because Nautilus doesn't allow for transparent backgrounds.

Other solutions like Wallpapoz are slow and when switching between desktops it takes too long to switch WP.

A team focused on researching support for new types of hardware devices should be started.

I find that linux developers in general always tend to wait for announced hardware, or until hardware becomes popular before adding general support for it. The issue is, that because most companies wont announce the existance of a new type of product until close to release date, we are always behind in support, because we never even have the infrastructure for it.

Some examples of technologies we should design an infrastructure for now are: Android mobiles, Multitouch, 3D printing, acellerometers and 3D projection. And there are plenty more which we should at least start coding proper libs for, so it doesn't break down into a bunch of adhoc drivers, and utilities that aren't universal.


Many people say "we should work on more important things". Like what? Making gnome look prettier? Thats all great, until manufacturers realise we don't have the infrastructure for the hardware they were thinking of designing, so they wont bother with us, because it will take months before programs could use it even with drivers (so they may as well just focus on the most popular platform).


Lets jump the gun, and start being the best at supporting NEWER hardware, and not just older hardware. If we don't do this, hardware in linux will never be truly "plug and play", instead, it will remain "plug and pray".

If we have an awesome infrastructure for cutting edge technologies, we will have something to brag about, and software developers will see linux as the best at supporting technologies in a generic manner.

The current Linux kernel cannot support hardware manufacturers for new devices out of the box. This is because the kernel coding team refuses to stabilise the API/ABI interfaces neccessary so that a single compiled module can work. So modules cannot continue to link.

At the moment, if you create a new device, if people want to support it during installation, the manufacturer must compile one for every minor kernel module release, and every distro (so maybe 200 different compiled modules), or the installer must carry gcc and compile it.

We should push the Linux kernel team to fix this, and bring it to attention. This would allow us to use the same compiled driver from 2.6.X all the way to the end of the series, instead of requiring it to be constantly updated. We could upgrade our kernels whenever we want, without worrying about recompiling our custom modules.

Think about it, a single kernel module which works on nearly any distro, and does not need constant updating/recompiling to remain working. Breaking modules constantly wont stop manufacturers like nvidia from keeping it closed source. All it will do is get rid of the annoying open source wrapper to their closed sourced module that needs constant recompiling.

Anyone can grab windows drivers, and load them using ndiswrapper (which linus says is a violation of the GPL), and the actual driver wont need updating for a new kernel, but ndiswrapper will. So if you have 10 wireless devices, with ndiswrapper, you only need to recompile 1 module (ndiswrapper), not 10. If the kernel was fixed, that number would be 0.

Jono Bacon posted a article on this a while ago:
http://www.linuxdevcenter.com/pub/a/linux/2004/09/02/driver_ease.html

Important: Before voting, make sure you know what you are voting for. Without this, it is IMPOSSIBLE for driver developers to provide drivers which are mostly guarenteed to work out of the box on their driver CD's, because they would need to compile something for every variation. We should be able to pop a CD in the drive, and load the same module anywhere. This would fix that. If you vote against, please say why. If you vote for, please say why. Just be honest, and ignore kernel politics, and base your decision on what you personally believe.

[....]

I had to generalize to keep the subject line short, however I'm noticing some very common trends on brainstorm as found on other internet forums (slashdot, digg, generic forums).

Especially via the comment system, we're rushing towards a scheme of developing "regulars" and people who dedicate themselves a bit too much to the preening of the system and micromanaging other peoples' ideas.
This is instead of simply allowing brainstorm to become an endless foundry for improvements.
Our comment format is starting to result in many people coming on trying to oppress ideas out of existence. No idea should have to suffer somebody's lack of sympathy or interest. It will eventually result in a gridlock of ideas as only the most mainstream will be promoted by cliques.

I don't know if this would require skilled moderation, applications for comment removal or if it would require an overall ***REDUCTION*** in functionality.

I myself have had people make the argument of:
o An idea not being valid because it could "take up mindshare" (which to me seems kind of vulgar in essence).
o Disagreement based on preference.
o Disagreement based on lack of endearment to the situation or criteria.
o Cited qualifications as a valid basis for reasoning.

This needs to be addressed lest we all be reduced to a mob of "citation needed" shouting snobs. I wouldn't be surprised if participation from both amateurs and professionals alike has been impacted by this sort of behavior if I'm encountering it now.

["I want my mom to feel welcome here"], but from what I've seen she'd be chased out by a bunch of neurotic self-appointed moderators.

Either way, we're headed for this on brainstorm and it would be foolish to think we're going to avoid the problem without trying.